Skip to main content

Full text of "Papers on the science of administration"

See other formats

immmmmmmm l.urwick 


Site (Sift nf 
Wylie Kilpatrick 





Digitized by the Internet Archive 

in 2011 with funding from 

LYRASIS Members and Sloan Foundation 








Edited by 




Copyright, 7937, by 
Institute of Public Administration 



The papers brought together in this collection are essays by men scientifically 
interested in the phenomena of administration. Most of these writers did their thinking 
independently, in some cases without any acquaintance with the others, or with their 
writings. The striking similarity and harmony of the analyses, nomenclature, and 
hypotheses, frequently set forth as principles, is thus doubly significant. 

Few of these papers have been published or publicly circulated in such a way as to 
make them accessible to practical administrators, scholars, or students. The immediate 
occasion for this publication is the fact that no copies of the essential papers in this col- 
lection could be found in any library in Washington at the time when the President's 
Committee on Administrative Management required these documents for the use of 
members of its research staff. 

In considering the proper content of this volume and in securing the permission of 
the authors, or of their executors and heirs, for publication, it was only natural to 
draw into the enterprise as associate editor L. Urwick. His contributions speak for 

It is the hope of the editors that the availability of these papers will advance the 
analysis of administration, assist in the development of a standard nomenclature, en- 
courage others to criticize the hypotheses with regard to administration herein set 
forth and to advance their own concepts fearlessly, and to point the way to areas 
greatly in need of exploration. If those who are concerned scientifically with the 
phenomena of getting things done through co-operative human endeavor will proceed 
along these lines, we may expect in time to construct a valid and accepted theory of 

This collection does not include anything from the writings of F. A. Cleveland, 
Charles A. Beard, and A. E. Buck, whose works are well known, unless it is in my own 
papers which grow directly from long years of association with them; nor from W. F. 
Willoughby, Leonard D. White, Marshall Dimock or John Gaus, all of whom have 
recently published books or essays dealing with administration, in a form easily accessi- 
ble to all. 

We are indebted to the authors, and to their heirs and agents, for permitting this 
publication of their pioneer works without expectation of royalty or reward, and to 
Miss Sarah Greer for the excellent translation of Henri Fayol's paper, The Administra- 
tive Theory in the State, which has not heretofore appeared in English. 

Administrators, scholars and students of the logic of administration who find this 
collection of value, must render their chief thanks to Miss Greer, who from her vantage 
point as librarian of the Institute of Public Administration, insisted that this collection 
was necessary, and who put back of the project that tireless pressure and devotion 
which is necessary to the accomplishment of any worthwhile purpose. 

L. G. 
New York City, June, 1937 



By Luther Gulick, Director, Institute of Public Administration; Eaton Professor of Munic- 
ipal Science and Administration, Columbia University; Member of the President's Committee 
on Administrative Management. 


By L. Urwick, O.B.E., M.C., M.A. Chairman, Urwick, On and Partners, Ltd.; Formerly 
Director of the International Management Institute, Geneva. 


By James D. Mooney, Vice-President, General Motors Corporation; Author {with Alan C. 
Reiley) of Onward Industry! 


By Henri Fayol, Management Specialist; Director of the Company of Commentry Four- 
chambault and Decazeville ; Member of the Committee of Collieries of France and of the Committee 
of the Iron Works of France; Member of the National Academy of Arts and Crafts; Officer of the 
Legion of Honor. 


With Special Reference to the Work of Henri Fayol 

By L. Urwick. 



By Henry S. Dennison, President, Dennison Manufacturing Company, Framingham, 
Massachusetts; Formerly Director, Service Relations, United States Post Office Department; 
Member of Advisory Committee, United States National Resources Committee. 


By L. J. Henderson, T. N. Whitehead and Elton Mayo, Graduate School of Business 
Administration, Harvard University, Boston, Massachusetts. 


By Mary Parker Follett, Author of The New State, and Creative Experience; Lecturer 
on Management, United States and England. 


By John Lee, C.B.E., Controller of the Central Telegraph Office, England, 1919-27; 
Lecturer at Industrial Conferences at Oxford and Cambridge; Author of various works on 
management; editor of the Dictionary of Industrial Administration. 


By V. A. Graicunas, Management. Consultant, Paris. 


By Luther Gulick. 


With Special Reference to 
Government in the United States 


Director, Institute of Public Administration 

Eaton Professor of Municipal Science and Administration 

Columbia University 



DECEMBER, 1 936 

Revised June, 1937 


Every large-scale or complicated enterprise requires many men to carry it for- 
ward. Wherever many men are thus working together the best results are secured 
when there is a division of work among these men. The theory of organization, there- 
fore, has to do with the structure of co-ordination imposed upon the work-division 
units of an enterprise. Hence it is not possible to determine how an activity is to be 
organized without, at the same time, considering how the work in question is to be 
divided. Work division is the foundation of organization; indeed, the reason for 

1 . The Division of Work 

It is appropriate at the outset of this discussion to consider the reasons for and the 
effect of the division of work. It is sufficient for our purpose to note the following 

Why Divide Work? 

Because men differ in nature, capacity and skill, and gain greatly in dexterity by 

Because the same man cannot be at two places at the same time; 

Because one man cannot do two things at the same time; 

Because the range of knowledge and skill is so great that a man cannot within his 
life-span know more than a small fraction of it. In other words, it is a question of 
human nature, time, and space. 

In a shoe factory it would be possible to have 1,000 men each assigned to making 
complete pairs of shoes. Each man would cut his leather, stamp in the eyelets, sew up 
the tops, sew on the bottoms, nail on the heels, put in the laces, and pack each pair in a 
box. It might take two days to do the job. One thousand men would make 500 pairs of 
shoes a day. It would also be possible to divide the work among these same men, using 
the identical hand methods, in an entirely different way. One group of men would be 
assigned to cut the leather, another to putting in the eyelets, another to stitching up the 
tops, another to sewing on the soles, another to nailing on the heels, another to insert- 
ing the laces and packing the pairs of shoes. We know from common sense and experi- 
ence that there are two great gains in this latter process: first, it makes possible the 
better utilization of the varying skills and aptitudes of the different workmen, and 
encourages the development of specialization; and second, it eliminates the time that 
is lost when a workman turns from a knife, to a punch, to a needle and awl, to a ham- 
mer, and moves from table to bench, to anvil, to stool. Without any pressure on the 
workers, they could probably turn out twice as many shoes in a single day. There 
would be additional economies, because inserting laces and packing could be assigned 
to unskilled and low-paid workers. Moreover, in the cutting of the leather there would 
be less spoilage because the less skillful pattern cutters would all be eliminated and 
assigned to other work. It would also be possible to cut a dozen shoe tops at the same 




time from the same pattern with little additional effort. All of these advances would 
follow, without the introduction of new labor saving machinery. 

The introduction of machinery accentuates the division of work. Even such a 
simple thing as a saw, a typewriter, or a transit requires increased specialization, and 
serves to divide workers into those who can and those who cannot use the particular 
instrument effectively. Division of work on the basis of the tools and machines used 
in work rests no doubt in part on aptitude, but primarily upon the development and 
maintenance of skill through continued manipulation. 

Specialized skills are developed not alone in connection with machines and tools. 
They evolve naturally from the materials handled, like wood, or cattle, or paint, or 
cement. They arise similarly in activities which center in a complicated series of inter- 
related concepts, principles, and techniques. These are most clearly recognized in the 
professions, particularly those based on the application of scientific knowledge, as in 
engineering, medicine, and chemistry. They are none the less equally present in law, 
ministry, teaching, accountancy, navigation, aviation, and other fields. 

The nature of these subdivisions is essentially pragmatic, in spite of the fact that 
there is an element of logic underlying them. They are therefore subject to a gradual 
evolution with the advance of science, the invention of new machines, the progress of 
technology and the change of the social system. In the last analysis, however, they 
appear to be based upon differences in individual human beings. But it is not to be 
concluded that the apparent stability of "human nature," whatever that may be, 
limits the probable development of specialization. The situation is quite the reverse. 
As each field of knowledge and work is advanced, constituting a continually larger and 
more complicated nexus of related principles, practices and skills, any individual will 
be less and less able to encompass it and maintain intimate knowledge and facility 
over the entire area, and there will thus arise a more minute specialization because 
knowledge and skill advance while man stands still. Division of work and inte- 
grated organization are the bootstraps by which mankind lifts itself in the process of 

The Limits of Division 

There are three clear limitations beyond which the division of work cannot to 
advantage go. The first is practical and arises from the volume of work involved in 
man-hours. Nothing is gained by subdividing work if that further subdivision results 
in setting up a task which requires less than the full time of one man. This is too obvi- 
ous to need demonstration. The only exception arises where space interferes, and in 
such cases the part-time expert must fill in his spare time at other tasks, so that as a 
matter of fact a new combination is introduced. 

The second limitation arises from technology and custom at a given time and 
place. In some areas nothing would be gained by separating undertaking from the cus- 
tody and cleaning of churches, because by custom the sexton is the undertaker; in build- 
ing construction it is extraordinarily difficult to re-divide certain aspects of electrical 
and plumbing work and to combine them in a more effective way, because of the juris- 
dictional conflicts of craft unions; and it is clearly impracticable to establish a division 
of cost accounting in a field in which no technique of costing has yet been developed. 


This second limitation is obviously elastic. It may be changed by invention and by 
education. If this were not the fact, we should face a static division of labor. It should be 
noted, however, that a marked change has two dangers. It greatly restricts the labor 
market from which workers may be drawn and greatly lessens the opportunities open 
to those who are trained for the particular specialization. 

The third limitation is that the subdivision of work must not pass beyond physical 
division into organic division. It might seem far more efficient to have the front half of 
the cow in the pasture grazing and the rear half in the barn being milked all of the 
time, but this organic division would fail. Similarly there is no gain from splitting a 
single movement or gesture like licking an envelope, or tearing apart a series of inti- 
mately and intricately related activities. 

It may be said that there is in this an element of reasoning in a circle; that the test 
here applied as to whether an activity is organic or not is whether it is divisible or not 
— which is what we set out to define. This charge is true. It must be a pragmatic test. 
Does the division work out? Is something vital destroyed and lost? Does it bleed? 

The Whole and the Parts 

It is axiomatic that the whole is equal to the sum of its parts. But in dividing up 
any "whole," one must be certain that every part, including unseen elements and rela- 
tionships, is accounted for. The marble sand to which the Venus de Milo may be re- 
duced by a vandal does not equal the statue, though every last grain be preserved; nor 
is a thrush just so much feathers, bones, flesh and blood; nor a typewriter merely so 
much steel, glass, paint and rubber. Similarly a piece of work to be done cannot be 
subdivided into the obvious component parts without great danger that the central 
design, the operating relationships, the imprisoned idea, will be lost. 

A simple illustration will make this clear. One man can build a house. He can lay 
the foundation, cut the beams and boards, make the window frames and doors, lay the 
floors, raise the roof, plaster the walls, fit in the heating and water systems, install the 
electric wiring, hang the paper, and paint the structure. But if he did, most of the work 
would be done by hands unskilled in the work; much material would be spoiled, and 
the work would require many months of his time. On the other hand, the whole job of 
building the house might be divided among a group of men. One man could do the 
foundation, build the chimney, and plaster the walls; another could erect the frame, 
cut the timbers and the boards, raise the roof, and do all the carpentry; another all the 
plumbing; another all the paper hanging and painting; another all the electric wiring. 
But this would not make a house unless someone — an architect — made a plan for 
the house, so that each skilled worker could know what to do and when to do it. 

When one man builds a house alone he plans as he works; he decides what to do 
first and what next, that is, he "co-ordinates the work." When many men work to- 
gether to build a house this part of the work, the co-ordinating, must not be lost sight 

In the "division of the work" among the various skilled specialists, a specialist in 
planning and co-ordination must be sought as well. Otherwise, a great deal of time 
may be lost, workers may get in each other's way, material may not be on hand when 
needed, things may be done in the wrong order, and there may even be a difference of 


opinion as to where the various doors and windows are to go. It is self-evident that the 
more the work is subdivided, the greater is the danger of confusion, and the greater is 
the need of overall supervision and co-ordination. Co-ordination is not something that 
develops by accident. It must be won by intelligent, vigorous, persistent and organized 

2. The Co-ordination of Work 

If subdivision of work is inescapable, co-ordination becomes mandatory. There is, 
however, no one way to co-ordination. Experience shows that it may be achieved in 
two primary ways. These are: 

1 . By organization, that is, by interrelating the subdivisions of work by allotting them 
to men who are placed in a structure of authority, so that the work may be co- 
ordinated by orders of superiors to subordinates, reaching from the top to the 
bottom of the entire enterprise. 

2. By the dominance of an idea, that is, the development of intelligent singleness of 
purpose in the minds and wills of those who are working together as a group, so 
that each worker will of his own accord fit his task into the whole with skill and 

These two principles of co-ordination are not mutually exclusive, in fact, no 
enterprise is really effective without the extensive utilization of both. 

Size and time are the great limiting factors in the development of co-ordination. 
In a small project, the problem is not difficult; the structure of authority is simple, and 
the central purpose is real to every worker. In a large complicated enterprise, the 
organization becomes involved, the lines of authority tangled, and there is danger that 
the workers will forget that there is any central purpose, and so devote their best 
energies only to their own individual advancement and advantage. 

The interrelated elements of time and habit are extraordinarily important in 
co-ordination. Man is a creature of habit. When an enterprise is built up gradually 
from small beginnings the staff can be "broken in" step by step. And when difficulties 
develop, they can be ironed out, and the new method followed from that point on as a 
matter of habit, with the knowledge that that particular difficulty will not develop 
again. Routines may even be mastered by drill as they are in the army. When, how- 
ever, a large new enterprise must be set up or altered overnight, then the real difficul- 
ties of co-ordination make their appearance. The factor of habit, which is thus an 
important foundation of co-ordination when time is available, becomes a serious 
handicap when time is not available, that is, when change rules. The question of co- 
ordination therefore must be approached with different emphasis in small and in large 
enterprises; in simple and in complex situations; in stable and in new or changing 

Co-ordination Through Organization 

Organization as a way of co-ordination requires the establishment of a system of 
authority whereby the central purpose or objective of an enterprise is translated into 


reality through the combined efforts of many specialists, each working in his own field 
at a particular time and place. 

It is clear from long experience in human affairs that such a structure of authority 
requires not only many men at work in many places at selected times, but also a single 
directing executive authority. 1 The problem of organization thus becomes the prob- 
lem of building up between the executive at the center and the subdivisions of work on 
the periphery an effective network of communication and control. 

The following outline may serve further to define the problem: 

I. First Step: Define the job to be done, such as the furnishing of pure water to all of 
the people and industries within a given area at the lowest possible cost: 
II. Second Step: Provide a director to see that the objective is realized; 

III. Third Step: Determine the nature and number of individualized and specialized 
work units into which the job will have to be divided. As has been seen above, 
this subdivision depends partly upon the size of the job (no ultimate subdivision 
can generally be so small as to require less than the full time of one worker) and 
upon the status of technological and social development at a given time; 

IV. Fourth Step: Establish and perfect the structure of authority between the director 
and the ultimate work subdivisions. 

It is this fourth step which is the central concern of the theory of organization. It is 
the function of this organization (IV) to enable the director (II) to co-ordinate and 
energize all of the sub-divisions of work (III) so that the major objective (I) may be 
achieved efficiently. 

The Span of Control 

In this undertaking, we are confronted at the start by the inexorable limits of 
human nature. Just as the hand of man can span only a limited number of notes on the 
piano, so the mind and will of man can span but a limited number of immediate mana- 
gerial contacts. The problem has been discussed brilliantly by Graicunas in his paper 
included in this collection. The limit of control is partly a matter of the limits of 
knowledge, but even more is it a matter of the limits of time and of energy. As a result 
the executive of any enterprise can personally direct only a few persons. He must 
depend upon these to direct others, and upon them in turn to direct still others, until 
the last man in the organization is reached. 

This condition placed upon all human organization by the limits of the span of 
control obviously differs in different kinds of work and in organizations of different 
sizes. Where the work is of a routine, repetitive, measurable and homogeneous char- 
acter, one man can perhaps direct several score workers. This is particularly true 
when the workers are all in a single room. Where the work is diversified, qualitative, 
and particularly when the workers are scattered, one man can supervise only a few. 
This diversification, dispersion, and non-measurability is of course most evident at the 

1 I.e., when organization is the basis of co-ordination. Wherever the central executive authority is composed of several 
who exercise their functions jointly by majority vote, as on a board, this is from the standpoint of organization still a 
"single authority"; where the central executive is in reality composed of several men acting freely and independently, then 
organization cannot be said to be the basis of co-ordination; it is rather the dominance of an idea and falls under the 
second principle stated above. 


very top of any organization. It follows that the limitations imposed by the span of 
control are most evident at the top of an organization, directly under the executive 

But when we seek to determine how many immediate subordinates the director of 
an enterprise can effectively supervise, we enter a realm of experience which has not 
been brought under sufficient scientific study to furnish a final answer. Sir Ian Hamil- 
ton says, "The nearer we approach the supreme head of the whole organization, the 
more we ought to work towards groups of three; the closer we get to the foot of the 
whole organization (the Infantry of the Line), the more we work towards groups of 
six." 2 

The British Machinery of Government Committee of 1918 arrived at the conclu- 
sion that "The Cabinet should be small in number — preferably ten or, at most, 
twelve." 3 

Henri Fayol said "[In France] a minister has twenty assistants, where the Ad- 
ministrative Theory says that a manager at the head of a big undertaking should not 
have more than five or six." 4 

Graham Wallas expressed the opinion that the Cabinet should not be increased 
"beyond the number of ten or twelve at which organized oral discussion is most 
efficient." 5 

Leon Blum recommended for France a Prime Minister with a technical cabinet 
modelled after the British War Cabinet, which was composed of five members. 6 

It is not difficult to understand why there is this divergence of statement among 
authorities who are agreed on the fundamentals. It arises in part from the differences 
in the capacities and work habits of individual executives observed, and in part from 
the non-comparable character of the work covered. It would seem that insufficient 
attention has been devoted to three factors, first, the element of diversification of func- 
tion; second, the element of time; and third, the element of space. A chief of public 
works can deal effectively with more direct subordinates than can the general of the 
army, because all of his immediate subordinates in the department of public works 
will be in the general field of engineering, while in the army there will be many 
different elements, such as communications, chemistry, aviation, ordnance, motor- 
ized service, engineering, supply, transportation, etc., each with its own tech- 
nology. The element of time is also of great significance as has been indicated above. 
In a stable organization the chief executive can deal with more immediate subordinates 
than in a new or changing organization. Similarly, space influences the span of control. 
An organization located in one building can be supervised through more immediate 
subordinates than can the same organization if scattered in several cities. When 
scattered there is not only need for more supervision, and therefore more supervisory 
personnel, but also for a fewer number of contacts with the chief executive because of 

2 Sir Ian Hamilton, "The Soul and Body of an Army." Arnold, London, 1921, p. 230. 

3 Great Britain. Ministry of Reconstruction. Report of the Machinery of Government Committee. H. M. Stationery 
Office, London, 1918, p. 5. 

4 Henri Fayol, "The Administrative Theory in the State." Address before the Second International Congress of 
Administrative Science at Brussels, September 13, 1923. Paper IV in this collection. 

6 Graham Wallas, "The Great Society." Macmillan, London and New York, 1919, p. 264. 
8 Leon Blum, "La Reforme Gouvernementale." Grasset, Paris, 1918. Reprinted in 1936, p. 59. 


the increased difficulty faced by the chief executive in learning sufficient details about 
a far-flung organization to do an intelligent job. The failure to attach sufficient im- 
portance to these variables has served to limit the scientific validity of the statements 
which have been made that one man can supervise but three, or five, or eight, or 
twelve immediate subordinates. 

These considerations do not, however, dispose of the problem. They indicate 
rather the need for further research. But without further research we may conclude 
that the chief executive of an organization can deal with only a few immediate sub- 
ordinates; that this number is determined not only by the nature of the work, but also 
by the nature of the executive ; and that the number of immediate subordinates in a 
large, diversified and dispersed organization must be even less than in a homogeneous 
and unified organization to achieve the same measure of co-ordination. 

One Master 

From the earliest times it has been recognized that nothing but confusion arises 
under multiple command. "A man cannot serve two masters" was adduced as a theo- 
logical argument because it was already accepted as a principle of human relation in 
everyday life. In administration this is known as the principle of "unity of command." 7 
The principle may be stated as follows: A workman subject to orders from several 
superiors will be confused, inefficient, and irresponsible; a workman subject to orders 
from but one superior may be methodical, efficient, and responsible. Unity of com- 
mand thus refers to those who are commanded, not to those who issue the commands. 8 

The significance of this principle in the process of co-ordination and organization 
must not be lost sight of. In building a structure of co-ordination, it is often tempting 
to set up more than one boss for a man who is doing work which has more than one 
relationship. Even as great a philosopher of management as Taylor fell into this error 
in setting up separate foremen to deal with machinery, with materials, with speed, 
etc., each with the power of giving orders directly to the individual workman. 9 The 
rigid adherence to the principle of unity of command may have its absurdities ; these 
are, however, unimportant in comparison with the certainty of confusion, inefficiency 
and irresponsibility which arise from the violation of the principle. 

Technical Efficiency 

There are many aspects of the problem of securing technical efficiency. Most of 
these do not concern us here directly. They have been treated extensively by such 
authorities as Taylor, Dennison, and Kimball, and their implications for general 
organization by Fayol, Urwick, Mooney, and Reiley. There is, however, one efficiency 
concept which concerns us deeply in approaching the theory of organization. It is the 
principle of homogeneity. 

It has been observed by authorities in many fields that the efficiency of a group 
working together is directly related to the homogeneity of the work they are perform- 

7 Henri Fayol, "Industrial and General Administration." English translation by J. A. Coubrough. International 
Management Association, Geneva, 1 930. 

8 Fayol terms the latter "unity of direction." 

8 Frederick Winslow Taylor, "Shop Management." Harper and Brothers, New York and London, 1911, p. 99. 


ing, of the processes they are utilizing, and of the purposes which actuate them. From 
top to bottom, the group must be unified. It must work together. 

It follows from this (1) that any organizational structure which brings together in 
a single unit work divisions which are non-homogeneous in work, in technology, or in 
purpose will encounter the danger of friction and inefficiency; and (2) that a unit based 
on a given specialization cannot be given technical direction by a layman. 

In the realm of government it is not difficult to find many illustrations of the 
unsatisfactory results of non-homogeneous administrative combinations. It is gener- 
ally agreed that agricultural development and education cannot be administered by 
the same men who enforce pest and disease control, because the success of the former 
rests upon friendly co-operation and trust of the farmers, while the latter engenders 
resentment and suspicion. Similarly, activities like drug control established in protec- 
tion of the consumer do not find appropriate homes in departments dominated by the 
interests of the producer. In the larger cities and in states it has been found that hos- 
pitals cannot be so well administered by the health department directly as they can be 
when set up independently in a separate department, or at least in a bureau with ex- 
tensive autonomy, and it is generally agreed that public welfare administration and 
police administration require separation, as do public health administration and wel- 
fare administration, though both of these combinations may be found in successful 
operation under special conditions. No one would think of combining water supply 
and public education, or tax administration and public recreation. In every one of 
these cases, it will be seen that there is some element either of work to be done, or of 
the technology used, or of the end sought which is non-homogeneous. 

Another phase of the combination of incompatible functions in the same office 
may be found in the common American practice of appointing unqualified laymen and 
politicians to technical positions or to give technical direction to highly specialized 
services. As Dr. Frank J. Goodnow pointed out a generation ago, we are faced here by 
two heterogeneous functions, "politics" and "administration," the combination of 
which cannot be undertaken within the structure of the administration without pro- 
ducing inefficiency. 

Caveamus Expertum 

At this point a word of caution is necessary. The application of the principle 
of homogeneity has its pitfalls. Every highly trained technician, particularly in the 
learned professions, has a profound sense of omniscience and a great desire for com- 
plete independence in the service of society. When employed by government he knows 
exactly what the people need better than they do themselves, and he knows how to 
render this service. He tends to be utterly oblivious of all other needs, because, after 
all, is not his particular technology the road to salvation? Any restraint applied to him 
is "limitation of freedom," and any criticism "springs from ignorance and jealousy." 
Every budget increase he secures is "in the public interest," while every increase 
secured elsewhere is "a sheer waste." His efforts and maneuvers to expand are "public 
education" and "civic organization," while similar efforts by others are "propaganda" 
and "politics." 

Another trait of the expert is his tendency to assume knowledge and authority in 


fields in which he has no competence. In this particular, educators, lawyers, priests, 
admirals, doctors, scientists, engineers, accountants, merchants and bankers are all the 
same — having achieved technical competence or "success" in one field, they come to 
think this competence is a general quality detachable from the field and inherent in 
themselves. They step without embarrassment into other areas. They do not remember 
that the robes of authority of one kingdom confer no sovereignty in another; but that 
there they are merely a masquerade. 

The expert knows his "stuff." Society needs him, and must have him more and 
more as man's technical knowledge becomes more and more extensive. But history 
shows us that the common man is a better judge of his own needs in the long run than 
any cult of experts. Kings and ruling classes, priests and prophets, soldiers and lawyers, 
when permitted to rule rather than serve mankind, have in the end done more to 
check the advance of human welfare than they have to advance it. The true place of 
the expert is, as A.E. said so well, "on tap, not on top." The essential validity of de- 
mocracy rests upon this philosophy, for democracy is a way of government in which 
the common man is the final judge of what is good for him. 

Efficiency is one of the things that is good for him because it makes life richer and 
safer. That efficiency is to be secured more and more through the use of technical 
specialists. These specialists have no right to ask for, and must not be given freedom 
from supervisory control, but in establishing that control, a government which ignores 
the conditions of efficiency cannot expect to achieve efficiency. 

3. Organizational Patterns 

Organization Up or Down? 

One of the great sources of confusion in the discussion of the theory of organiza- 
tion is that some authorities work and think primarily from the top down, while others 
work and think from the bottom up. This is perfectly natural because some authorities 
are interested primarily in the executive and in the problems of central management, 
while others are interested primarily in individual services and activities. Those who 
work from the top down regard the organization as a system of subdividing the enter- 
prise under the chief executive, while those who work from the bottom up, look upon 
organization as a system of combining the individual units of work into aggregates 
which are in turn subordinated to the chief executive. It may be argued that either 
approach leads to a consideration of the entire problem, so that it is of no great signifi- 
cance which way the organization is viewed. Certainly it makes this very important 
practical difference: those who work from the top down must guard themselves from 
the danger of sacrificing the effectiveness of the individual services in their zeal to 
achieve a model structure at the top, while those who start from the bottom, must 
guard themselves from the danger of thwarting co-ordination in their eagerness to 
develop effective individual services. 

In any practical situation the problem of organization must be approached from 
both top and bottom. This is particularly true in the reorganization of a going concern. 
May it not be that this practical necessity is likewise the sound process theoretically? 
In that case one would develop the plan of an organization or reorganization both 


from the top downward and from the bottom upward, and would reconcile the two 
at the center. In planning the first subdivisions under the chief executive, the principle 
of the limitation of the span of control must apply; in building up the first aggregates 
of specialized functions, the principle of homogeneity must apply. If any enterprise 
has such an array of functions that the first subdivisions from the top down do not 
readily meet the first aggregations from the bottom up, then additional divisions and ad- 
ditional aggregates must be introduced, but at each further step there must be a less and 
less rigorous adherence to the two conflicting principles until their juncture is effected. 
An interesting illustration of this problem was encountered in the plans for the re- 
organization of the City of New York. The Charter Commission of 1934 approached 
the problem with the determination to cut down the number of departments and 
separate activities from some 60 to a manageable number. It was equally convinced 
after conferences with officials from the various city departments that the number 
could not be brought below 25 without bringing together as "departments" activities 
which had nothing in common or were in actual conflict. This was still too many for 
effective supervision by the chief executive. As a solution it was suggested by the author 
that the charter provide for the subdividing of the executive by the appointment of 
three or four assistant mayors to whom the mayor might assign parts of his task of 
broad supervision and co-ordination. Under the plan the assistant mayors would bring 
all novel and important matters to the mayor for decision, and through continual inti- 
mate relationship know the temper of his mind on all matters, and thus be able to re- 
lieve him of great masses of detail without in any way injecting themselves into the 
determination of policy. Under such a plan one assistant mayor might be assigned to 
give general direction to agencies as diverse as police, parks, hospitals, and docks with- 
out violating the principle of homogeneity any more than is the case by bringing these 
activities under the mayor himself, which is after all a paramount necessity under a 
democratically controlled government. This is not a violation of the principle of homo- 
geneity provided the assistant mayors keep out of the technology of the services and de- 
vote themselves to the broad aspects of administration and co-ordination, as would the 
mayor himself. The assistants were conceived of as parts of the mayoralty, not as parts 
of the service departments. That is, they represented not the apex of a structure built 
from the bottom up, but rather the base of a structure extended from the top down, the 
object of which was to multiply by four the points of effective contact between the 
executive and the service departments. 10 

Organizing the Executive 

The effect of the suggestion presented above is to organize and institutionalize the 
executive function as such so that it may be more adequate in a complicated situation. 
This is in reality not a new idea. We do not, for example, expect the chief executive 
to write his own letters. We give him a private secretary, who is part of his office and 
assists him to do this part of his job. This secretary is not a part of any department, he 

10 This recommendation was also presented to the Thatcher Charter Commission of 1935, to which the author was a 
consultant. A first step in this direction was taken in Sec. 9, Chap. I of the new charter which provides for a deputy 
mayor, and for such other assistance as may be provided by ordinance. 


is a subdivision of the executive himself. In just this way, though on a different plane, 
other phases of the job of the chief executive may be organized. 

Before doing this, however, it is necessary to have a clear picture of the job itself. 
This brings us directly to the question, "What is the work of the chief executive? What 
does he do?" 

The answer is POSDCORB. 

POSDCORB is, of course, a made-up word designed to call attention to the various 
functional elements of the work of a chief executive because "administration" and 
"management" have lost all specific content. 11 POSDCORB is made up of the initials 
and stands for the following activities: 

Planning, that is working out in broad outline the things that need to be done 
and the methods for doing them to accomplish the purpose set for the 

Organizing, that is the establishment of the formal structure of authority 
through which work subdivisions are arranged, defined and co-ordinated 
for the defined objective; 

Staffing, that is the whole personnel function of bringing in and training the 
staff and maintaining favorable conditions of work; 

Directing, that is the continuous task of making decisions and embodying 
them in specific and general orders and instructions and serving as the 
leader of the enterprise; 

Co-ordinating, that is the all important duty of interrelating the various 
parts of the work; 

Reporting, that is keeping those to whom the executive is responsible in- 
formed as to what is going on, which thus includes keeping himself and 
his subordinates informed through records, research and inspection; 

Budgeting, with all that goes with budgeting in the form of fiscal planning, 
accounting and control. 

This statement of the work of a chief executive is adapted from the functional 
analysis elaborated by Henri Fayol in his "Industrial and General Administration." It 
is believed that those who know administration intimately will find in this analysis a 
valid and helpful pattern, into which can be fitted each of the major activities and 
duties of any chief executive. 

If these seven elements may be accepted as the major duties of the chief executive, 
it follows that they may be separately organized as subdivisions of the executive. The 
need for such subdivision depends entirely on the size and complexity of the enterprise. 
In the largest enterprises, particularly where the chief executive is as a matter of fact 
unable to do the work that is thrown upon him, it may be presumed that one or more 
parts of POSDCORB should be suborganized. 

It is interesting to note that this has been recognized in many of our larger govern- 
mental units, though there has been until recently no very clear philosophy lying back 

11 See Minutes of the Princeton Conference on Training for the Public Service, 1935, p. 35. See also criticism of this 
analysis in Lewis Meriam, "Public Service and Special Training," University of Chicago Press, 1936, pp. 1, 2, 10 and 15, 
where this functional analysis is misinterpreted as a statement of qualifications for appointment. 


of the arrangements which have been made. For example, in the federal government 
at the present time one may identify the separate institutionalization of: 

Planning, under the National Resources Committee, though as yet rudi- 
mentary in development; 

Staffing, under the Civil Service Commission, though it has missed its con- 
structive role; 

Reporting, under the National Emergency Council and the Central Statistics 
Board, in elementary form. 

Budgeting, under the Budget Bureau; 

Each of these agencies is as a matter of fact now serving as a managerial arm of 
the chief executive, particularly budgeting and planning. It will be observed that 
directing, co-ordinating and organizing are not institutionalized, but remain undiffer- 
entiated and unimplemented in the hands of the President. In view of the fact that he 
is swamped now, it would seem desirable to take out the organizing function, turning 
that over to the efficiency research division of the Budget Bureau, and then to increase 
the immediate personal staff of the White House with five or six high grade personal 
assistants to the President, along the line planned for the City of New York, to make it 
possible for him to deal adequately with the directing and co-ordinating functions. 12 
What is most needed is a group of able and informed men who will see that the Presi- 
dent has before him all relevant facts and that all appropriate clearance is secured 
before decisions are made and that a decision once made is known to those who are 

In view of the fact that the job of the President as Chief Executive is POSDCORB, 
institutionalization must not be allowed to take any one of these functions out of his 
office. They are and must remain parts of him. This was recognized when the Budget 
Bureau was set up in 1921, though it is only in the past three years that the office has 
actually functioned as an arm of the Chief Executive. The Civil Service Commission 
was similarly set up directly under the President, but the fact that it has an unwieldy 
triple head and has been interested primarily in the negative aspects of personnel has 
prevented its true development. Each one of these managerial establishments should 
be a part of the executive office. Their budgets should be brought together, and as far 
as possible they should be in the White House itself. It should be as natural and easy 
for the President to turn to the chairman of the planning board, or to the civil service 
commissioner when confronted by problems in their fields as it is now to call in the 
budget director before deciding a matter of finance. 

A further illuminating illustration of the subdivision of the executive function is 
found in the Tennessee Valley Authority. The executive head of the authority is a board 
of three. The members of the board are also individually directors of various sections 
of the work. There is no general manager. Organizing, staffing, reporting, and budget- 
ing are all separately institutionalized under the board, which deals with planning 
itself. Directing is in part handled by the board and in part by its individual members. 
To preserve some measure of co-ordination, the board has appointed a co-ordinator as 

12 As recommended by the President to Congress on January 12, 1937, in accordance with the program of the Presi- 
dent's Committee on Administrative Management. 


one of its chief officers. The results of this type of organization will be watched with 
interest by students of administration. 13 

Aggregating the Work Units 

In building the organization from the bottom up we are confronted by the task 
of analyzing everything that has to be done and determining in what grouping it can 
be placed without violating the principle of homogeneity. This is not a simple matter, 
either practically or theoretically. It will be found that each worker in each position 
must be characterized by: 

1 . The major purpose he is serving, such as furnishing water, controlling crime, or 
conducting education; 

2. The process he is using, such as engineering, medicine, carpentry, stenography, 
statistics, accounting; 

3. The persons or things dealt with or served, such as immigrants, veterans, Indians, 
forests, mines, parks, orphans, farmers, automobiles, or the poor; 

4. The place where he renders his service, such as Hawaii, Boston, Washington, the 
Dust Bowl, Alabama, or Central High School. 

Where two men are doing exactly the same work in the same way for the same 
people at the same place, then the specifications of their jobs will be the same under 
1, 2, 3, and 4. All such workers may be easily combined in a single aggregate and super- 
vised together. Their work is homogeneous. But when any of the four items differ, then 
there must be a selection among the items to determine which shall be given preced- 
ence in determining what is and what is not homogeneous and therefore combinable. 

A few illustrations may serve to point the problem. Within the City of New York, 
what shall be done with the doctor who spends all of his time in the public schools 
examining and attending to children in the Bronx? Shall we (1) say that he is primarily 
working for the school system, and therefore place him under the department of 
education? (2) say that he is a medical man, and that we will have all physicians in the 
department of health? (3) say that he is working with children, and that he should 
therefore be in a youth administration? or (4) say that he is working in the Bronx and 
must therefore be attached to the Bronx borough president's office? Whichever answer 
we give will ignore one or the other of the four elements characterizing his work. The 
same problem arises with the lawyer serving the street construction gang on damage 
cases in Brooklyn, the engineer who is working for the department of health in Rich- 
mond, and the accountant examining vouchers and records in Queens for the district 

Departments Vertical and Horizontal 

The nature of the interrelation between departments organized on the basis of 
purpose and those organized on the basis of process may be illustrated best by considering 
the former as vertical departments, and the latter as horizontal departments. This idea 
is presented pictorially in Chart I. In this chart four sample city departments are 
presented vertically, each broken down into its obvious specialized activities and 

13 Organization altered by the establishment of the office of General Manager after this was written. 



workers. Each one of these is conceived of as a department set up to perform an im- 
portant service, and fully equipped with all of the staff which it needs for the accom- 
plishment of the major purpose involved. For example, the department of health, as 
here outlined, is established to guard and improve the health of all of the people of a 
given city. In this work it will need not only doctors, bacteriologists, nurses, and in- 
spectors, but also lawyers, engineers, janitors, repair men, architects, clerks, statisti- 
cians, and budget, accounting, purchasing, and personnel staffs. The same is true of 
each of the other departments listed. In each case, it will be seen, there is (1) a central 
core of management which must be equipped with enthusiasm, experience, ability, 
knowledge of the purpose to be accomplished, and general acquaintance with the 
technologies used; (2) a group of workers peculiar to the department (such as fire 
fighters in the fire department, and school teachers in the department of education) ; 
and (3) a considerable number of highly important skilled or professional workers who 
are common to all, or to several departments. 

These latter workers, because they are common to all or to several departments, 
and because they are engaged in the same sort of work group by group, may be 
brought together in appropriate departments. These are the "process" departments, 
and are shown on the accompanying chart as horizontal units, outlined in red. Turning 
the page on its side, and looking at the red network, it will be noted that each depart- 
ment is made up of (1) a central core of management which must be equipped with 
enthusiasm, experience, ability, broad understanding of the general purposes of the 
government and its various services, and extensive knowledge of the technology in- 
volved; (2) a group of workers peculiar to the department, but with the same basic 
training as the remainder of the department (such as the staff engaged in tax adminis- 
tration in the department of finance); and (3) a group of skilled and professional 
workers working in other departments on assignment (like a lawyer assigned by the 
corporation counsel to the department of education), or performing services required 
to complete the job of another department (like park patrol, when handled by the 
police department). 

Geographical division, that is departmentalization on the basis of place, presents 
an analogous alternative plan of organization. This too may be illustrated by a dia- 
gram, as is done in Chart II. Here again, the vertical departments, the black network, 
represent the great service departments, organized to perform a city-wide task. 
Instead of showing their technical work subdivision as on the previous chart, attention 
is here directed to their regional offices. It is assumed that each of the departments in 
this illustration maintains local offices within the same areas because of the natural 
divisions of the city. Such a situation is found in some of the largest cities because sec- 
tions of the city are sharply defined by rivers, by parks, or by other clear demarcations. 
If each of the regional offices were split off from the vertical departments in question 
and joined with the other local offices in the same area, then new departments would 
appear, as is shown by the horizontal green network. Under this plan, there would be 
in each area a regional office which contained within itself health, education, police, 
recreation, and other services under a co-ordinating head. This co-ordinating head 
would be a regional assistant mayor. It would be his job to see that all of the city serv- 
ices touching the people within his area fitted together and made sense, and that any 




Assistant Directors 



Assistant Superintendents 



Assistant Chiefs 



Assistant Commissioners 

< < 

H o « 
52 53" 

Private secretaries 
File clerks 

Private secretaries 
Tile clerks 

Private secretaries 
File clerks 

Private secretaries 
File clerks 


Budget officer 
Purchasing officer 

Budget officer 
Purchasing officer 

Budget officer 
Purchasing officer 

Budget officer 

a th 
Purchasing officer £ 5 


Personnel manager 

Personnel manager 

Personnel manager 

Personnel manager 

S Eh 


Repair force 


Repair force 

Repair force 

Landscape staff 
Repair force 



Switchboard operator 




Laboratory assistants 


Classroom teacherB 
Special teachers 
Recreation leaders 
Playground supervisors 

Traffic supervisor 

Switchboard operator 

Crime laboratory staff 
Police school staff 

Uniformed force 
Detective force 
Traffic force 
Jail staff 
Mounted force 

Communications staff 

Plant laboratory staff 


Recreation leaders 
Playground supervisors 
Park police 

Traffic force 

Zoo staff 
Switchboard operator 




K (^ 


O l* 




Motorized service 

Motorized service 


J K 

Motorized service 

Motorized service 

Black network — Purpose departments 
Red network — Process departments 





I— I 















r ^ 

r \r a 

r 1 

^ ' 


<H CO 





«K O 
CO *H 









+» Cm «h 






CO «h <h 











CO 43 



■P rj fl 

co g 

CD "ffl 

cd a 




5-1 CO C O -HO 

•H O 


TO O £ -H CO «H 
3 -rl O W> W) 

lO -H 

•H -H 





CO hO 



& g 43 co .d © 

•d .□. ti Pi -P « 

.d co 





if rt 

■p rt 




TOO g 







co co 





M Eh P 






«H CD 






Vi O 
TO *H 









-P «H «H 






CO Cm <h 











CO -P 


cal S 








•H O 

•rl O 


tO -H 

tO -rl 

•H -rt 




d-d W 

id ^ d pej 

TO g 

. M 

_ «) 

to l\0 


,d CD 

d © 

+3 t? 




•p Pd 

•2 PS 










co co 






W eh 













43 «H 



ra "a 


CO 43 




cal S 




P iH 




O W 



4> CD 









W Si 





+» «H 


w 'a 



CO -P 


Pi tO 









<d J3 











40 CO 










43 CD 








•H O 
lO iH 






•H O 









•H O 

























© a 
>d o 

■H iH 

co w 
•p ft) 





















a. o 

Ph P< 

a o 




J V 

so Mao 


aais keqos, 




aais xssm 



Black network — Purpose departments 
Red network — Process departments 
Green network — Regional departments 


new problem was given prompt attention. If the problem proved to be local, he would 
see that it was dealt with by one of his own little departments, but if it proved to be 
city-wide in scope, then he would see that it was passed along for city-wide attention. 
No diagram is presented here of departments organized on the basis of clientele or 
materiel, the former are similar in nature to the purpose departments, unless the 
clientele is geographically restricted in which case they follow the pattern of Chart II. 
The materiel departments are more akin to the process departments shown by the red 
network of Chart I, in fact the motorized service shown at the bottom of the chart is 
such a grouping. 

The Tangled Fabric 

The effort is made in Chart III to show how departments established on various 
principles are woven together to form a single fabric. The black network, the red net- 
work and the green network of Charts I and II are repeated. This might better have 
been done in three dimensions, 14 but it will suffice to portray the complexity of the 
problem involved. 

In Chart III no special nomenclature is included as the entire structure is hypo- 
thetical. The first red bar may be considered as the finance functions; the second, 
statistical services; the third, medical services; and the last, motorized service. The 
green network, inserted merely for illustration, might be the public information and 
complaint offices, serving not only several of the service departments, but also the tax 
assessment bureau of the department of finance. 

It is a further function of this chart to show how certain departments are in part 
vertical and in part horizontal. This is true, for example, of the health department 
where that department is set up with a reasonably full staff to perform all of the public 
health functions, and then does some work for other departments through its profes- 
sional staff. 

Practice and Theory 

These charts serve to bring out the major questions and considerations which 
arise in the practice and theory of organization. It may not be amiss therefore to note 
certain of them in this connection, though they are referred to more systematically 

1. Is there any advantage in placing specialized services like private secretaries or 
filing, in the red network? In a very small organization, yes; in a large organization, 
no. In a small organization, where there is not a full-time job each day, and more 
than a full-time job on some days for each secretary, it is better to have a central 
secretarial pool than to have a private secretary for each man. In a large organiza- 
tion, the reverse is true, and there is great loss from failure to accumulate specialized 
knowledge and experience. The grouping of all stenographers, moreover, adds 
little or nothing to co-ordination. 

2. The engineers are another matter. Engineering is highly specialized within itself, 
different engineers dealing with design, construction, operation, hydraulics, elec- 

11 Cf. "State Organization Model" and "Jungle Gym" by Henry Toll, 1934. 


tricity, sanitation, cement, highways, bridges, steel, etc. Only a very large volume 
and scope of engineering service can possibly afford to have even a modest array of 
specialized talent of this sort. Therefore the small department is doomed to the 
employment of a staff, which though expert in one line will be relatively inexpert 
in some other. When all of the engineering work of the city is brought together in 
a single department, this danger is greatly reduced. Hence, engineering should be 
brought together except where departments are so large that they can by them- 
selves afford a sufficient volume of work to use the most efficient specialization 
which the current technology makes possible. 

The establishment of horizontal technical process departments such as engineering, 
statistics, or motorized service, along with departments set up vertically on the 
basis of major purpose does introduce a tendency to the co-ordination of these 
phases of their work, and by a sort of inductance, to a greater co-ordination of 
other phases of their work as well. 

3. The major arms of co-ordination are found, however, in the horizontal services 
which handle planning, budget, and personnel. In this context the budget includes 
not alone the making of fiscal plans, but also their execution, together with the 
research, accounting, inspection and informational services which are involved. 
The major reason for bringing these services together in the red network is not 
increased efficiency through specialization, but rather the development of tools of 

4. From the standpoint of co-ordination it would seem that the heads of departments 
should visualize and understand the network in which they belong, so that they 
may proceed intelligently and energetically not only to carry out the work under 
their direction, but to see that that part of their work which falls within the over- 
lapping purpose, or process, or regional area of another department may be fitted 
together harmoniously. Such an approach may serve to dispel the feeling of ex- 
clusive ownership and jealousy so common in government departments. 

5. The chief executive, by the same token, should understand the network of the 
organization and should make a special endeavor to watch the areas of overlap 
and friction, and should hold the co-ordinating elements of the red network not 
only in his own hands, but in proper check so that they may not interfere with the 
accomplishment of the major purposes of the government. 

With these illustrations and comments in mind we may now turn to the consider- 
ation of the advantages and disadvantages which may be expected with the applica- 
tion of one or another of the principles of departmentalization. Unfortunately we must 
rest our discussion primarily on limited observation and common sense, because little 
scientific research has been carried on in this field of administration. 

Organization by Major Purpose >/' 

Organization by major purpose, such as water supply, crime control, or educa- 
tion, serves to bring together in a single large department all of those who are at work 
endeavoring to render a particular service. Under such a policy, the department of ed- 
ucation will contain not only teachers and school administrators, but also architects, 


engineers, chauffeurs, auto mechanics, electricians, carpenters, janitors, gardeners, 
nurses, doctors, lawyers, and accountants. Everything that has to do with the schools 
would be included, extending perhaps even to the control of traffic about school 
properties. Similarly the department of water supply would include not only engineers 
and maintenance gangs, but also planners, statisticians, lawyers, architects, account- 
ants, meter readers, bacteriologists, and public health experts. 

The advantages of this type of organization are three: first, it makes more certain 
the accomplishment of any given broad purpose or project by bringing the whole job 
under a single director with immediate control of all the experts, agencies and services 
which are required in the performance of the work. No one can interfere. The director 
does not have to wait for others, nor negotiate for their help and co-operation; nor ap- 
peal to the chief executive to untangle a conflict. He can devote all his energies to get- 
ting on with the job. 

Second, from the standpoint of self-government, organization by purpose seems 
to conform best to the objectives of government as they are recognized and understood 
by the public. The public sees the end result, and cannot understand the methodology. 
It can therefore express its approval or disapproval with less confusion and more effec- 
tiveness regarding major purposes than it can regarding the processes. 

Third, it apparently serves as the best basis for eliciting the energies and loyalties 
of the personnel and for giving a focus and central drive to the whole activity, because 
purpose is understandable by the entire personnel down to the last clerk and inspector. 

The statement of these strong points of organization by major purpose points the 
way to its dangers. These are to be found, first, in the impossibility of cleanly dividing 
all of the work of any government into a few such major purposes which do not overlap 
extensively. For example, education overlaps immediately with health and with rec- 
reation, as does public works with law enforcement. The strong internal co-ordination 
and drive tends to precipitate extensive and serious external conflict and confusion, just 
as there is more danger of accident with a high powered motor car. This is apparent 
particularly in the development of a reasonable city plan, or in arriving at a consistent 
policy throughout the departments for the maintenance of properties, or in handling 
legal matters, or arranging similar work and salary conditions. The lawyers, engineers, 
accountants, doctors of different departments will all have their own ideas as to how 
similar matters are to be dealt with. 

Second, there is danger that an organization erected on the basis of purpose will 
fail to make use of the most up-to-date technical devices and specialists because the 
dominance of purpose generally tends to obscure the element of process, and because 
there may not be enough work of a given technical sort to permit efficient subdivision. 

Third, there is also danger in such an organization that subordinate parts of the 
work will be unduly suppressed or lost sight of because of the singleness of purpose, 
enthusiasm and drive of the head of the department. For example, medical work with 
children when established under the department of education as a division is likely to 
receive less encouragement than it would if independently established in the health de- 
partment, because after all the department of education is primarily interested in 
schools and has its own great needs and problems. 

Fourth, a department established on the basis of purpose falls easily into the habit 


of overcentralization, and thus fails to fit its service effectively to the people. Or if it 
does decentralize its services, as do the fire department, the police department, the 
health department and the department of education of New York City, the representa- 
tives of these departments in the field do not always make the best use of each other's 
assistance and co-operation, and when any difficulty does arise, it is such a long way to 
the top where co-ordination can be worked out, that it is easier to get along without it. 
Fifth, an organization fully equipped from top to bottom with all of the direct and 
collateral services required for the accomplishment of its central purpose, without the 
need of any assistance from other departments, drifts very easily into an attitude and 
position of complete independence from all other activities and even from democratic 
control itself. 

Organization by Major Process 

Organization by major process, such as engineering, teaching, the law, or medi- 
cine, tends to bring together in a single department all of those who are at work making 
use of a given special skill or technology, or are members of a given profession. Under 
such a policy the department of law would comprise all of the lawyers and law clerks, 
including those who are devoting their time to school matters, or water supply suits, or 
drafting ordinances. The department of engineering and public works would have all 
the engineers, including those concerned with planning, design, construction, main- 
tenance and other phases of engineering work, wherever that work was found. This 
would include the work in the parks, on the streets, in the schools, and in connection 
with water, sewer, waste and other services. The department of health would include 
all of the doctors, nurses, and bacteriologists, and would not only carry on the general 
public health work, but would do the medical and nursing work for the schools, the 
water department, the department of social welfare, etc., as has been outlined above. 

In every one of these cases it will be observed that the basis of organization is the 
bringing together in a single office or department of all the workers who are using some 
particular kind of skill, knowledge, machinery, or profession. This principle of organ- 
ization has the following advantages: 

First, it guarantees the maximum utilization of up-to-date technical skill and by 
bringing together in a single office a large amount of each kind of work (technologi- 
cally measured), makes it possible in each case to make use of the most effective divi- 
sions of work and specialization. 

Second, it makes possible also the economies of the maximum use of labor saving 
machinery and mass production. These economies arise not from the total mass of the 
work to be performed, not from the fact that the work performed serves the same gen- 
eral purpose but from the fact that the work is performed with the same machine, with 
the same technique, with the same motions. For example, economy in printing comes 
from skill in typesetting, printing, and binding and the use of modern equipment. It 
makes no difference to the printer whether he is printing a pamphlet for the schools, a 
report for the police department, or a form for the comptroller. Unit costs, efficiency 
in the doing of the job, rest upon the process, not the purpose. 16 

16 Of course overall efficiency by the same token rests on the purpose, not the process. For example, a report may be 
printed at a phenomenally low cost, but if the pamphlet has no purpose, the whole thing is a waste of effort. 


Third, organization by process encourages co-ordination in all of the technical 
and skilled work of the enterprise, because all of those engaged in any field are brought 
together under the same supervision, instead of being scattered in several departments 
as is the case when organization is based upon some other principle. 

Fourth, it furnishes an excellent approach to the development of central co- 
ordination and control when certain of the services such as budgeting, accounting, 
purchasing, and planning are set up on a process basis and used as instruments of 
integration even where other activities are set up on some other basis. 

Fifth, organization by process is best adapted to the development of career 
service, and the stimulation of professional standards and pride. A career ladder can 
be erected very much more easily in a department which is from top to bottom engi- 
neers, or doctors, or statisticians, or clerks, than it can in a department which is partly 
engineers, partly doctors, partly statisticians, partly clerks. In the vertical depart- 
ments, 16 the rungs of a professional ladder are a flying trapeze requiring the employee 
in his upward course to swing from department to department. This cannot be ac- 
complished "with the greatest of ease." 

These are the major advantages of organization on the basis of process. There are, 
of course, offsetting difficulties. As in the case of any other principle of organization, it 
is impossible to aggregate all of the work of the government on such a basis alone. It is 
not difficult to do so for engineering and medicine and teaching, but it becomes im- 
possible when we reach typing and clerical work. It cannot furnish a satisfactory basis 
for doing the whole job in any large or complicated enterprise. 

In the second place, there is always the danger that organization by process will 
hinder the accomplishment of major purposes, because the process departments may 
be more interested in how things are done than in what is accomplished. For example, a 
housing department which must clear the slums, build new low cost tenements and 
manage them, and inspect existing housing and approve new building plans, may 
find it difficult to make rapid progress if it must draw its legal help from the corpora- 
tion counsel, its architects from the department of engineering, its enforcement 
officers from the police department, and its plans from the planning commission, 
particularly if one or more of these departments regards public housing as a nuisance 
and passing fad. There are also accountants who think that the only reason for the 
running of a government is the keeping of the books ! 

Third, experience seems to indicate that a department built around a given 
profession or skill tends to show a greater degree of arrogance and unwillingness to 
accept democratic control. This is perhaps a natural outgrowth of the insolence of 
professionalism to which reference has already been made. 

Fourth, organization by process is perhaps less favorable to the development of a 
separate administrative service, because it tends to bring rather narrow professional 
specialists to the top of each department, men who are thereby disqualified for transfer 
to administrative posts in other fields. 

And finally, the necessity of effective co-ordination is greatly increased. Purpose 
departments must be co-ordinated so that they will not conflict but will work shoulder 
to shoulder. But whether they do, or do not, the individual major purposes will be ac- 

16 Sec Chart I. 


complished to a considerable extent and a failure in any service is limited in its effect 
to that service. Process departments must be co-ordinated not only to prevent con- 
flict, but also to guarantee positive co-operation. They work hand in hand. They must 
also time their work so that it will fit together, a factor of lesser significance in the pur- 
pose departments. A failure in one process affects the whole enterprise, and a failure to 
co-ordinate one process division, may destroy the effectiveness of all of the work that 
is being done. 

While organization by process thus puts great efficiency within our reach, this 
efficiency cannot be realized unless the compensating structure of co-ordination is 

Organization by Clientele or Materiel 

Organization on the basis of the persons served or dealt with, or on the basis of 
the things dealt with, tends to bring together in a single department, regardless of the 
purpose of the service, or the techniques used, all of those who are working with a 
given group or a given set of things. Examples are the veterans' administration which 
deals with all of the problems of the veteran, be they in health, in hospitals, in insur- 
ance, in welfare, or in education; and the immigration bureau which deals with 
immigrants at all points, including legal, financial, and medical services. Depart- 
mentalization on the basis of materiel is more common in private business than in 
public. Department stores, for example, have separate departments for furniture, 
hardware, drugs, jewelry, shoes, etc., and have separate buyers and sales forces for 
each. In many communities the school is in reality such a service, as it concentrates 
most of the community services touching children in school, including medical inspec- 
tion, corrective treatment, free lunches and recreation, and certain phases of juvenile 
crime. The Forest Service is another organization based on materiel — in this case, trees. 

The great advantage of this type of organization is the simplification and co- 
ordination of the service of government in its contact with the consumer. He does not 
encounter first one and then another representative, each of whom does or demands 
something different or even something contradictory. At the international border one 
may be met by an immigration inspector who is interested in one's nationality and 
residence, by a customs inspector who is interested in goods brought in, by an agri- 
cultural inspector interested in certain pests, by a game warden interested in guns and 
rods, etc. In New York City, at one time, each tenement was subject to separate in- 
spection at periodic intervals by men interested in slums, crime, fire escapes, plumbing, 
fire hazards, and electric wiring. Many cases have been reported of conflicting orders 
being issued to individual property owners under such a system. 

A second advantage is found in the increasing skill which attends the handling 
over and over of the same material. 

A third gain arises from the elimination of duplicate travel, particularly in dealing 
with widely separated or sparsely distributed work. 

The disadvantages of an organization which brings together all of the contacts 
with a given individual or thing are: 

First, it tends to sacrifice the efficiency of specialization, because it must after all 
perform several otherwise specialized functions through the same organization or even 


at times through the same agent. For example, if the effort is made to combine all 
building inspections in the field, the same man must examine plumbing, wiring, living 
conditions and fire escapes. Or, at the international border, the same man must know 
and enforce the immigration law, the customs laws, the hunting laws, and the crop 
pest laws, and must possess the required special knowledge in the various technical 
fields involved. Or, in the veterans' administration the same director must supervise 
and direct specialists in medicine, institutional administration, insurance, and voca- 
tional education and rehabilitation. 

It is to be noted that this difficulty has been overcome in certain notable instances 
by the creation of a new specialist profession to deal with the specific combination of 
functions brought together in a given service. The best illustration is the United 
States Forest Service. This can be done only where there is a large volume of work 
and an opportunity for a career within the service itself. 

A second difficulty is found in the impossibility of applying the principle of divi- 
sion by persons served to all of the work of a government, without encountering ex- 
tensive conflict and duplication. It is not difficult to pick out special groups like the 
aged, the youth, the criminal, the veteran, the real estate owner, etc., but when all is 
said and done there remains a great number of the ordinary citizens that does not fall 
into any single grouping. Each individual will appear in various groups at various times, 
and in the general group known as "the public" the rest of the time. And it is clearly 
impossible to organize a special department for the public, with all of the heteroge- 
neous elements which this would entail from the standpoint of dissimilar technologies 
and conflicting objectives. It must be remembered also that even such departments as 
seem to be organized on the basis of persons served do not as a matter of fact cover all 
of the services rendered to or government contacts with a class of individuals. Taking 
even the most expanded school systems where the schools really are youth depart- 
ments, it will be found that general quarantine is enforced by the health authorities, 
that children benefit also by police traffic and crime prevention work, that they drink 
water furnished by the water department, and milk protected by the division of foods, 
and live in homes supervised by the housing authority, and are protected by the fire 

A third difficulty arises from the danger of dominance by favor-seeking pressure 
groups. Departments set up by clientele seldom escape political dominance by those 
groups, and are generally found to be special pleaders for those groups, at times in 
opposition to the general interest of society as a whole. This is in part due to the fact 
that the organization itself is often brought into being through the action of a pressure 
group and its demand for a special agency to serve it, but it is also continued through 
the efforts of the agency once established to marshal and maintain a group in its 
support. It follows that agencies so set up as to maintain or develop their own pressure 
backing are peculiarly difficult of democratic control and tend not to fit into a co- 
ordinated social policy. 

Organization by Place 

Organization on the basis of the place at which the service is performed brings 
together all of those who work in a limited area regardless of the service they are per- 


forming or of the techniques they represent. This is the general practice in territorial 
or colonial governments. Even where the home government has separate departments 
interested in health, education, law enforcement, natural resources, labor relations, 
and commercial development, it will be found that these departments have no direct 
representatives within a given territory or colony, but that there is a single representa- 
tive of the home government there, under whom there are organized separate divisions 
to deal with health, education, law enforcement, natural resources, etc. In other 
words all or most of the representatives of the home government in the area are 
brought together in a single local agency, in place of serving as the far distant field 
representatives of the various central departments. 

It is not only in colonial government that this plan is utilized. It is found also in 
greater or lesser degree in many of the regular departments of all large governmental 
units. In some of the largest city police systems the city is divided into precincts, and 
most of the police activities within a given area are under the complete direction of a 
precinct officer through whom all communication to and from headquarters must go. 
A similar situation is found in the postal system. Though the Washington office con- 
tains divisions in charge of mail delivery, stamp sale, postal savings, money orders, 
parcel post and other services, these services do not each maintain independent local 
offices. A single local office is maintained with a postmaster in charge, under whom 
there are separate divisions for performing these particular services and for reporting 
the results of their work to the several supervisory and control divisions at head- 
quarters. The Treasury Department, on the other hand, though it has customs officers, 
income tax examiners, bank examiners, narcotic officers, secret service, and others in 
the field representing various administrative divisions of the Treasury in Washington, 
does not establish a general local director of these field services, but supervises each in 
turn directly from Washington, or through some regional office which has no organic 
relationship with any one of the other services. 

The most extreme cases of subdivision of the work of government are found in the 
American states where the state has not only sub-divided itself into geographic areas 
for the performance of certain types of governmental service, but has actually turned 
over to these geographic areas a large measure of the right to determine how the 
local service shall be conducted. This is known as "home rule," and is found in all 
kinds of local government, particularly in the cities. It is common in the conduct of 
schools, the management of the police, the enforcement of justice, and the mainte- 
nance of the courts, all of which are legally "state functions." It is also the general 
basis of operation for poor relief, local highways, water supply, waste removal, property 
tax administration, and health administration. Where these functions are turned over 
to cities, villages, towns, or counties, they are in reality sub-organized on a geographi- 
cal basis. It is significant that in France, Napoleon called his major geographical sub- 
divisions Departements. From the standpoint of the theory of organization, this is a 
thoroughly accurate designation. 

It should be noted that every department in every government of any size must be 
broken down geographically. In no other way can it reach the people who are to be 
served or who are to be controlled. In the government of the United States, only 20 per 
cent of the employees are located in Washington, and thousands of this number are 


I. No Geographical 

All lingers 

II. Centralized Geographical 

Short Arms 
Long Fingers 

The Field 

III. Decentralized Geo- 
graphical Subdivisions 

Long Arms 
Short Fingers 

The Field 


actually in the field. Eighty per cent are regularly assigned to offices or work scattered 
throughout the country. They carry the government to the people and to the soil. 
The same is true of state governments and of city governments. Only a small part of 
the force actually works in the state capitol or the city hall. The real work of govern- 
ment is done out among the people in the various sections of the state or city. In the 
supervision of these forces it is often necessary to establish some form of regional 
organization, if for no other reason than to save the time of the supervisory officers who 
cannot be in two places at once. It is thus generally a question as to how high up in the 
organization geographical subdivision shall be introduced. Obviously this may be 
done at the very top, as the first division of the work under the chief executive of an 
enterprise, or it may be introduced far down the line after the major divisions have 
been set up by purpose, by process, or by clientele. The former may be termed primary 
geographical subdivision and the latter secondary, tertiary, or subordinate geographical 
subdivision. A department or major activity, like the Tennessee Valley Authority, 
which is set up on the basis of geographical boundaries is in fact a primary geographi- 
cal subdivision of the government. 

It is important to note also that geographical subdivisions of an office may be 
within the central office or they may be located away from the central office in the 
field, that is, they may be centralized or decentralized. In the first case the central 
office would itself have an "Eastern Division," a "Southern Division," a "Western 
Division," etc., each with an extensive staff at headquarters. In the United States 
Department of State, the Division of Latin American Affairs and the Division of Near 
Eastern Affairs, to mention but two, are illustrations of this system of organization. 
The alternative is the establishment of the regional suboffices in the field, far removed 
from headquarters. Under this system the "Eastern Division" would have an office 
within the Eastern area, and would have its staff located there. An illustration of this is 
found in the Tennessee Valley Authority and in the plans for the seven power authori- 
ties suggested by Senator Norris and endorsed by the President in his Message of 
June 3, 1937. It is, with certain modifications, the underlying plan of the Federal 
Reserve System and of the Farm Credit Administration. 

The essential difference between these types of geographical division of work may 
be illustrated best by Chart IV. 

These diagrams are oversimplified for purposes of illustration. They do not, for 
example, distinguish between the purpose and the process activities, nor do they deal 
with the structure in the field for the co-ordination of the various fingers. 17 Then, too, 
the whole chart is spread out to gain separation of the lines, whereas the true picture is 
more like a piano duet in which the treble of one player overlaps the base clef of the 
other, and in which the score is far from distinct, and "swing" is the rule ! 

The advantages of departmentalization on the basis of geographical areas, that 
is on the basis of superior geographical subdivision, are fairly obvious in practice. 
They consist first of the greater ease of co-ordination of services rendered and controls 
exercised within a given area; second, of the greater tendency to adapt the total pro- 
gram to the needs of the areas served, not alone because of the discretion resting 
within the divisions, but also because the needs and differences of the areas will be 

17 See page 36. 


more vigorously represented at headquarters in the general consideration of broad 
policy; and third, of the greater ease with which co-operative relations may be estab- 
lished with subordinate governmental units, which are of necessity first of all geo- 
graphically defined units. Decentralization of geographical divisions strengthens these 
tendencies, and serves, moreover, to reduce travel costs, short circuit adjustment 
problems, cut red tape, and speed up all joint activities and administrative decisions. 
It increases not only the awareness of the officials to local needs and to the interrelation 
of service and planning problems, but develops a new sensitivity to the process of 
democratic control through intimate association of the officials with the people 

With decentralized subdivision a large amount of discretion must be delegated to 
the men in charge of field offices; in fact, they must be men of ability equal to if not 
superior to those who would be selected to head centralized departments of similar 

The difficulties of primary geographic subdivision are also not far to seek. They 
consist of the increased difficulty of maintaining a uniform nation-wide, state-wide, or 
city-wide policy; the danger of too narrow and short-sighted management; and the 
increased difficulty of making full use of technical services and the highest specializa- 
tion because of the division of the work into limited blocks. Decentralization tends to 
enhance these difficulties by reason of physical isolation. It introduces other factors as 
well, such as higher costs for supervisory personnel, the general hesitancy of central 
administrative heads to delegate sufficient real power, the lesser prestige of localized 
officials, and the increased tendency of such a system to come under the control of 
localized logrolling pressure groups. Political parties under our system of representa- 
tion are based upon geographical areas. An administrative system also set up by areas 
is peculiarly subject to spoliation by politicians as long as we have the spoils system. 

Whenever the concept of geographic areas is introduced into the structure of 
organization, either as a primary or as a subordinate plan of division of work, there is 
always the further practical problem of delineating appropriate boundaries. This is 
particularly difficult when it is planned to deal with several activities widely differing 
in their nature and technology. There is always the danger that the tasks to be dealt 
with do not follow compact geographic boundaries and that the administrative separa- 
tion introduced by the geographic division will complicate rather than simplify the 

Line and Staff 

The army has contributed much to the theory of organization. Not the least of 
these contributions has been the concept of line and staff. The implications of this idea 
for other than military organizations have been discussed brilliantly by Urwick. by 
Frederick A. Cleveland, and by Mooney and Reiley. 18 In view of the clarity of these 
expositions it is astonishing that so many who discuss administration fail to understand 
what a staff is and what function it performs in the organization. 

18 L. Urwick, "Organization as a Technical Problem," Paper II of this collection; Frederick A. Cleveland, "Expert 
Staff Aids to Management," Industrial Service and Equipment Co., Boston, 1918; James D. Mooney and Alan C. Reiley, 
"Onward Industry!" Harper and Brothers, New York, 1931. 


When the work of government is subjected to the dichotomy of "line" and 
"staff," there are included in staff all of those persons who devote their time ex- 
clusively to the knowing, thinking and planning functions, and in the line all of the 
remainder who are, thus, chiefly concerned with the doing functions. The overhead 
directing authority of the staff group, usually a board or committee, is the "general 

Obviously those in the line are also thinking and planning, and making sugges- 
tions to superior officers. They cannot operate otherwise. But this does not make them 
staff officers. Those also in the staff are doing something; they do not merely sit and 
twiddle their thumbs. But they do not organize others, they do not direct or appoint 
personnel, they do not issue commands, they do not take responsibility for the job. 
Everything they suggest is referred up, not down, and is carried out, if at all, on the 
responsibility and under the direction of a line officer. 

The important point of confusion in considering line and staff has arisen in speak- 
ing of the budget director, the purchasing agent, the controller, the public relations 
secretary, as "staff" officers. On the basis of the definition it is clear that they are all 
line officers. They have important duties of direction and control. When administra- 
tive responsibility and power are added to any staff function, that function thereby 
becomes immediately and completely a line function. There is no middle ground. 

The chief value of the line and staff classification is to point to the need (1) of 
developing an independent planning agency as an aid to the chief executive, and (2) 
of refusing to inject any element of administrative authority and control into such an 

The necessity for central purchase, for personnel administration, for budgeting 
and for fiscal control rests on other considerations and not on the philosophy of the 
general staff. 

4. Interrelation of Systems of Departmentalization 

Students of administration have long sought a single principle of effective depart- 
mentalization just as alchemists sought the philosophers' stone. 19 But they have sought 
in vain. There is apparently no one most effective system of departmentalism. 

Each of the four basic systems of organization is intimately related with the other 
three, because in any enterprise all four elements are present in the doing of the work 
and are embodied in every individual workman. Each member of the enterprise is 
working for some major purpose, uses some process, deals with some persons, and 
serves or works at some place. 

If an organization is erected about any one of these four characteristics of work, 
it becomes immediately necessary to recognize the other characteristics in constructing 
the secondary and tertiary divisions of the work. For example, a government which is 
first divided on the basis of place will, in each geographical department, find it neces- 
sary to divide by purpose, by process, by clientele, or even again by place; and one 

19 Charles A. Beard, "The Administration and Politics of Tokyo." Macmillan, New York, 1923, Ch. 3; A. E. Buck, 
"Administrative Consolidation in State Governments," 5th ed. National Municipal League, New York, 1930; Great 
Britain, Ministry of Reconstruction, Report of the Machinery of Government Committee. H. M. Stationery Office, 
London, 1918; Luther Gulick, "Principles of Administration," National Municipal Review, vol. 14, July, 1925, pp. 400- 
403; W. F. Willoughby, "Principles of Public Administration." Johns Hopkins Press, Baltimore, 1927, Part I, Ch. 5. 


divided in the first instance by purpose, may well be divided next by process and then 
by place. While the first or primary division of any enterprise is of very great signifi- 
cance, it must none the less be said that there is no one most effective pattern for deter- 
mining the priority and order for the introduction of these interdependent principles. 
It will depend in any case upon the results which are desired at a given time and place. 

An organization is a living and dynamic entity. Each activity is born, has its 
periods of experimental development, of vigorous and stable activity, and, in some 
cases, of decline. A principle of organization appropriate at one stage may not be 
appropriate at all during a succeeding stage, particularly in view of the different 
elements of strength and of weakness which we have seen to exist in the various systems 
of departmentalization. In any government various parts of its work will always 
stand at different stages of their life cycle. It will therefore be found that not all of the 
activities of any government may be appropriately departmentalized neatly on the 
basis of a single universal plan. Time is an essential element in the formula. 

Another variable is technological development. The invention of machines, the 
advance of applied science, the rise of new specializations and professions, changes in 
society and in the way men work and move in their private life must be continually 
reflected in the work of government, and therefore in the structure of government. 
Medieval governments made use of warriors, priests, artists, builders, and tax gath- 
erers; they had no place for sanitary engineers, chemists, entomologists, pneumatic 
drill operators and typists. Before you organize a statistical division there must be 
statistical machinery and statistical science, but as soon as there are such machinery 
and science, any large organization which fails to recognize the fact in its organization 
may greatly lessen its utilization of the newly available tools and skills. 

A further variable influencing the structure of any enterprise is its size, measured 
not so much by the amount of work done as by the number of men at work and their 
geographical dispersion. A drug store is an excellent illustration of the problem en- 
countered. It must have a prescription department with a licensed pharmacist, no 
matter how small it is, because of the technological requirements involved. But it does 
not need to have a separate medicine and supply department, refreshment depart- 
ment, book department, toy department, sporting goods department, cigar depart- 
ment, and delivery department, each with a trained manager, buyer and sales force, 
unless it is a big store. In the small store, the pharmacist may even be the manager, the 
soda jerker, and the book dispenser. If the business is big enough, it may be desirable 
to have more than one store in order to reach the customers, thus introducing geo- 
graphical subdivision. Similarly, in government the nature of the organization must 
be adapted not only to the technological requirements but also to the size of the under- 
taking and the dispersion of its work. 

Measurements and Organization 

We must not fail to recognize the importance of the techniques of administration 
upon what is possible and what is not possible in the organization of large-scale enter- 
prise. How far could men go in the organization of business without double-entry 
bookkeeping and the balance sheet? How far could they go without the invention of the 
corporation? What has been the influence upon the practicable size and structure of 


business of such things as the telephone, the elevator, the adding and computing ma- 
chines, or the conveyor belt? It is not difficult, looking into the past, to note these indi- 
vidual inventions and to trace their influence upon the structure of organization. 
Just as changes in production and distribution have altered the work requirements of 
every enterprise, so these inventions and technological advances in administration have 
altered the process of central co-ordination, direction and control. 

If we turn to the future, we are thus compelled to see that systems of organization 
which we now find to be necessary to produce specific results may become completely 
archaic and unnecessary with the invention of new administrative machines and tech- 
niques. Of very great importance, for example, is the effort now being made to find 
"measurements of administration" in many fields. 20 Though little advance has been 
made as yet, the high perfection to which automatic and machine accounting has 
been brought, offers real promise. If, for example, effective units of highway work are 
worked out by the experts, these, with the aid of rapid machine statistics, can give an 
executive effective control over a farflung decentralized organization of a character 
which he could not handle under existing technologies. If to this is added the new 
developments of communication, with two-way radio, and even television, we 
may find that we are confronted with entirely new possibilities in the field of organi- 
zation. 21 

Structure and Co-ordination 

The major purpose of organization is co-ordination, as has been pointed out 
above. It should therefore be noted that each of the four principles of departmentaliza- 
tion plays a different role in co-ordination. In each case the highest degree of co-ordi- 
nation takes place within the departments set up, and the greatest lack of co-ordination 
and danger of friction occurs between the departments, or at the points where they 

If all of the departments are set up on the basis of purpose, then the task of the 
chief executive in the field of co-ordination will be to see that the major purposes are 
not in conflict and that the various processes which are used are consistent, and that 
the government as it touches classes of citizens or reaches areas of the community is 
appropriate, rational, and effective. He will not have to concern himself with co-ordi- 
nation within the departments, as each department head will look after this. 

If all of the departments are set up on the basis of process, the work methods will 
be well standardized on professional lines, and the chief executive will have to see that 
these are co-ordinated and timed to produce the results and render the services for 
which the government exists, and that the service rendered actually fits the needs of 
the persons or areas served. 

20 A. E. Buck, "Measuring the Results of Government," National Municipal Review, vol. 13, March, 1924, pp. 152-157; 
Luther Gulick, "Wanted: A Measuring Stick for School Systems," National Municipal Review, vol. 18, January, 1929, 
pp. 3-5; Clarence E. Ridley, "Measuring Municipal Government," Municipal Administration Service, New York, 1927. 
See also articles in Public Management, 1937, by Clarence E. Ridley and H. A. Simon; John J. Theobald, "An Economic 
Analysis of Highway Administration in the State of New York," pages 140-192, and 657-679 of the Report of the New 
York State Commission for the Revision of the Tax Laws, 1935. 

21 An excellent illustration of this is already upon us in police administration, where motorized service and two-way 
radio have made precinct organization and small town set-ups of doubtful value. 


If place be the basis of departmentalization, that is, if the services be decentral- 
ized, then the task of the chief executive is not to see that the activities are co-ordinated 
locally and fit the locality, but to see that each of these services makes use of the standard 
techniques and that the work in each area is part of a general program and policy. 

If the work of the government be departmentalized in part on the basis of pur- 
pose, in part on the basis of process, in part on the basis of clientele, and in part on the 
basis of place, it will be seen that the problems of co-ordination and smooth operation 
are multiplied and that the task of the executive is increased. Moreover, the nature of 
his work is altered. In an organization in which all of the major divisions follow one 
philosophy, the executive himself must furnish the interdepartmental co-ordination 
and see that things do not fall between two stools. In an organization built on two or 
more bases of departmentalization, the executive may use, for example, the process 
departments as a routine means of co-ordinating the purpose departments. None the 
less the task of the executive is extraordinarily complicated. There is also great danger 
in such an organization that one department may fail to aid or actually proceed to 
obstruct another department. When departments cross each other at right angles, the 
danger of collision is far greater and far more serious than when their contacts are 
along parallel lines at their respective outer limits. 

The Holding Company Idea 

A large enterprise engaged in many complicated activities which do not require 
extensive or intimate co-ordination may need only the loosest type of central co- 
ordinating authority. Under such conditions, each activity may be set up, on a purpose 
basis, as virtually independent, and the central structure of authority may be nothing 
more than a holding company. In practice various industrial holding companies, 
particularly in the power field, require little or no co-ordination whatsoever. They 
have no operating services in common, and seem to have few interrelations except in 
finance. It has been suggested that the larger governmental units are in a comparable 
position, and that they may well be looked upon not as single enterprises like the Ford 
Motor Company, but rather as if they were each holding companies like the American 
Telephone and Telegraph Company, or General Motors. From this point of view, the 
government of the United States would be the parent company, and each department 
would be an independent subsidiary. While the parent company would give certain 
central services and require conformity to certain central plans and policies, each 
subsidiary, that is each department, would be given extensive freedom to carry on as it 
saw fit, and the President at the center of the parent company would not pretend to do 
more than prevent conflict and competition. 

This point of view is helpful to the student of administration in that it brings out 
two important factors: 

1. It makes clear the important difference between the operating functions and de- 
partments, such as agriculture, war, and labor, and the co-ordinating and central 
services, such as the budget, planning, and personnel. In the holding company 
analogy, the former would be subsidiaries, while the latter would be functions of 
the parent company; and 

2. It directs attention to the kind of service to which the central agencies, including 


the President and the Cabinet, should limit themselves in any case. If the co-ordinat- 
ing agencies of the government would look upon themselves as holding company 
officials and staff, they would devote their energies to the larger problems of co- 
ordination, and would leave to the departments and their staffs the internal prob- 
lems of operation. 

While this attitude toward the respective functions of the operating and the co- 
ordinating services of the government may be valuable for certain purposes, it cannot 
be accepted as the sound theoretical foundation for the consideration of the federal 
government, or of any of the governments of the states or larger cities. It is not a 
satisfactory analogy for four important reasons: 

1 . There is but one board of directors in the governmental set-up, and a single avenue 
of democratic responsibility; 

2. The interrelations between the various departments are many and intimate, re- 
quiring extensive and continuous co-ordination; 

3. In government there must be highly developed uniform standards and methods, 
particularly in finance and personnel; and 

4. There is in government no simple, final measure of successful operation of sub- 
sidiaries like the profit and loss statement in business. Supervisory relations must be 
intimate and complete, not distant and limited. 

In the actual operation of the larger American governmental units we are, as a 
matter of fact, confronted at the same time by too much activity by the co-ordinating 
authorities, and by too little co-ordination. This anomalous situation seems to come 
about because of the lack of understanding both by experts and by laymen of the true 
function of the chief executive; because of the lack of proper managerial staffs attached 
to the chief executive; and because of the tendency of legislative bodies to step over the 
line into administration and to meddle with appointments. It must be recognized that 
the chief executive of any enterprise has but a limited amount of time and energy at his 
command. These he can use either in participating in detail in the administration of a 
few activities, or in dealing broadly with the policies and problems of many activities. 
If the task of the executive is first of all co-ordination, it would seem that the latter is 
his true function. But in any large enterprise, the executive cannot perform this func- 
tion intelligently or skillfully unless he has adequate assistance. Where he is denied 
such assistance, he must act either tardily or ruthlessly, and an executive who recoils 
from either course is immediately drawn down into the minutiae of administration 
and fails to perform his main job. 

In public administration the holding company concept is helpful if it is used to 
emphasize the need of broad co-ordination and the methods of achieving it. It must be 
recognized, however, that government is actually not a holding company at all. 

Other Means of Interdepartmental Co-ordination 

In the discussion thus far it has been assumed that the normal method of inter- 
departmental co-ordination is hierarchical in its operation. That is, if trouble develops 
between a field representative (X) of one department and the field representative (A) 
of another department, that the solution will be found by carrying the matter up the 
line from inferior to superior until the complaint of Mr. X and the complaint of Mr. A 



finally reach their common superior, be he mayor, governor or President. In actual 
practice, there are. also other means of interdepartmental co-ordination which must be 
regarded as part of the organization as such. Among these must be included planning 
boards and committees, interdepartmental committees, co-ordinators, and officially 
arranged regional meetings, etc. These are all organizational devices for bringing 
about the co-ordination of the work of government. Co-ordination of this type is 
essential. It greatly lessens the military stiffness and red tape of the strictly hierarchical 
structure. It greatly increases the consultative process in administration. It must be 
recognized, however, that it is to be used only to deal with abnormal situations and 
where matters of policy are involved, as in planning. The organization itself should be 
set up so that it can dispose of the routine work without such devices, because these 
devices are too dilatory, irresponsible and time-consuming for normal administration. 
Wherever an organization needs continual resort to special co-ordinating devices in 
the discharge of its regular work, this is proof that the organization is bad. These 
special agencies of co-ordination draw their sanction from the hierarchical structure 
and should receive the particular attention of the executive authority. They should not 
be set up and forgotten, ignored, or permitted to assume an independent status. 

The establishment of special regional co-ordinators to bring about co-operation 
in a given region between the local representatives of several central agencies presents 
special difficulties. There are three chief plans which have been tried. One is the device 
of propinquity, the juxtaposition of offices in the same building or city, and reliance on 
ordinary daily contact. Another is the establishment of a loose committee, or "confer- 
ence," meeting locally from time to time to discuss local problems of co-ordination, 
under a local chairman who is actually nothing but a presiding officer and formulator 








Field Servioe 







.Looal Division 





"1 1 1 

of agenda. A third plan is the establishment of such groups under the chairmanship of 
a regional co-ordinator designated by and responsible to a central office of co-ordina- 
tion of field service. When this is attempted, to whom should the head of the field 
service be subordinate? For example, in the accompanying diagram to whom should 
X report? Certainly not to A, or to B, or to C. The central director of a co-ordination 
field service must be on a par with the central directors of the services which are being 
co-ordinated in the field, or possibly even on a higher plane. 


In the devices of co-ordination, one must recognize also joint service contracts 
and coincident personnel appointments. Independent agencies may be pulled to- 
gether in operation through such use of the same staff or service. There are many illus- 
trations of the former, especially in engineering services. The county agent who is at the 
same time a county, a state, and a federal official is an example of the latter. 

A great obstacle in the way of all of these plans of co-ordination is found in the 
danger of introducing confusion in direction through the violation of the principle of 
unity of command, and also in the difference in the level of authority of those who are 
brought together in any interdepartmental or intergovernmental co-ordinating ar- 
rangement. The representatives of the Department of Agriculture, for example, may 
have a large measure of responsibility and power, and may therefore be in a position 
to work out an adjustment of program through conference and to agree to a new line 
of conduct, but the representatives of the Army, coming from an entirely different 
kind of an organization, are sure to be in a position where they cannot make any ad- 
justments without passing the decision back to headquarters. 

5. Co-ordination by Ideas 

Any large and complicated enterprise would be incapable of effective operation if 
reliance for co-ordination were placed in organization alone. Organization is neces- 
sary; in a large enterprise it is essential, but it does not take the place of a dominant 
central idea as the foundation of action and self-co-ordination in the daily operation 
of all of the parts of the enterprise. Accordingly, the most difficult task of the chief ex- 
ecutive is not command, it is leadership, that is, the development of the desire and will 
to work together for a purpose in the minds of those who are associated in any activity. 

Human beings are compounded of cogitation and emotion and do not function 
well when treated as though they were merely cogs in motion. Their capacity for great 
and productive labor, creative co-operative work, and loyal self-sacrifice knows no 
limits provided the whole man, body-mind-and-spirit, is thrown into the program. 


After all that has been written about morale in war and in work, the psychology of 
group effort, and the art of leadership, it is not necessary to elaborate this point here. 
It is appropriate, however, to note the following specific elements which bear directly 
upon the problem of co-ordination: 

1. Personnel administration becomes of extraordinary significance, not merely from 
the standpoint of finding qualified appointees for the various positions, but even 
more from the standpoint of assisting in the selection of individuals and in the 
maintenance of conditions which will serve to create a foundation of loyalty and 

The new drive for career government service and for in-service training derives its 
significance not so much from the fact that better persons will enter the service 
when the chance for promotion is held out to them, but from the fact that a career 
service is a growing and learning service, one that believes in the work and in the 
future of the enterprise. 


2. Even where the structure of the organization is arranged to produce co-ordination 
by authority, and certainly in those realms in which the structure as such is wanting, 
the effort should be made to develop the driving ideas by co-operative effort and 
compromise so that there may be an understanding of the program, a sense of par- 
ticipation in its formulation, and enthusiasm in its realization. 

3. Proper reporting on the results of the work of the departments and of the govern- 
ment as a whole to the public and to the controlling legislative body, and public ap- 
preciation of good service rendered by public employees is essential, not merely as 
a part of the process of democratic control, but also as a means to the development 
of service morale. 

4. As a matter of public policy the government should encourage the development of 
professional associations among the employees of the government, in recognition of 
the fact that such associations can assist powerfully in the development of standards 
and ideals. In situations where it is natural, office and shop committees should be 
built up. 

5. A developing organization must be continually engaged in research bearing upon 
the major technical and policy problems encountered, and upon the efficiency of 
the processes of work. In both types of research, but particularly in the latter, mem- 
bers of the staff at every level should be led to participate in the inquiries and in the 
development of solutions. 

6. There is need for a national system of honor awards which may be conspicuously 
conferred upon men and women who render distinguished and faithful, though not 
necessarily highly advertised public service. 

7. The structure of any organization must reflect not only the logic of the work to be 
done, but also the special aptitudes of the particular human beings who are brought 
together in the organization to carry through a particular project. It is the men 
and not the organization chart that do the work. 

Dominant Ideals 

The power of an idea to serve as the foundation of co-ordination is so great that 
one may observe many examples of co-ordination even in the absence of any single 
leader or of any framework of authority. The best illustration is perhaps a nation at 
war. Every element steps into line and swings into high gear "to help win the war." 
The co-ordination is enthusiastic and complete, within the limits of knowledge of 
course. In an old stable community, small enough for each person to know the other, 
even competing businesses generally work along together in harmony. The town 
board, the school board, the park commission, the overseer of the poor, though answer- 
able to no single executive, manage to get along with each other and each to fit his 
part of the work into that of the others to arrive at a sensible result for the whole pic- 
ture. Men of intelligence and good will find little difficulty in working together for a 
given purpose even without an organization. They do not need to be held in line or 
driven to do a specific task in a specific way at a specific time. They carry on because 
of their inner compulsion, and may in the end accomplish a far better result for that 
very reason. 


Hire and Fire 

Closely akin to these considerations is the place of discipline in the technique of 
co-ordination. In most American thinking on management it has been assumed that 
the power to hire and fire must of necessity follow the hierarchical lines of the organi- 
zation. Even students of administration have assumed that a chart of any organization 
may be drawn in terms either of the lines of appointment or in terms of the duties per- 
formed, because it was thought that the two are coincident, or should be. American 
business men have shared the same view. In their labor relations they take the tradi- 
tional position that no man can run a business unless he has the free and unlimited 
power to hire and fire. It is the cornerstone of their thinking. 

From the standpoint of modern management, however, competent observers have 
come to see that this approach is neither justified by the facts nor conducive to effective 
results. In practical situations employers do not as a matter of fact have the power to 
fire, even when they have the legal right. There are public relations, and increasingly 
there are labor unions. Frequently employers do not have the unlimited power to 
appoint. Here again there are standards, public relations, and closed shops. An indi- 
vidual minor executive in a large corporation generally has his employees selected for 
him. He is not free either to hire or to fire. These limitations of the right to hire and 
fire are destined, it is clear, to become more and more restrictive. 

It becomes increasingly clear, therefore, that the task of the administrator must be 
accomplished less and less by coercion and discipline and more and more by persua- 
sion. In other words, management of the future must look more to leadership and less 
to authority as the primary means of co-ordination. 

These undoubted facts of experience are re-emphasized here not for the purpose 
of arguing that the consideration of the philosophy of organization is after all a "fools' 
contest," but rather to insist that the more important and the more difficult part of 
co-ordination is to be sought not through systems of authority, but through ideas and 
persuasion, and to make clear the point that the absurdities of the hierarchical system 
of authority are made sweet and reasonable through unity of purpose. It may well be 
that the system of organization, the structure of authority, is primarily important in 
co-ordination because it makes it easy to deal with the routine affairs, and thereby 
lessens the strain placed upon leadership, so that it can thus devote itself more fully to 
the supreme task of developing consent, participation, loyalty, enthusiasm, and 
creative devotion. 

6. Co-ordination and Change 

The Limits of Co-ordination 

Are there limits to co-ordination? Is mankind capable of undertaking activities 
which though interrelated are beyond man's power of systematic co-ordination? 

In the field of big business, economists and lawyers have raised this question, and 
have asked whether some of the larger corporations were not already too large for 
effective operation. Certainly in government, we face what often seem superhuman 
tasks because of their size and complexity. And not too encouraging are such efforts as 
we have seen in Italy, Germany, and Russia, where government is co-ordinating not 
only the social controls but also production and distribution. In a lesser degree, and 


with greater reliance upon voluntary action within a broad framework of regulation, 
England, France and the United States are also striving for the same ends, with 
greater and greater imposition of public controls. There are apparently no limits to the 
effort mankind is prepared to make to render life more secure and more abundant 
through socially enforced co-ordination. If there are limits to man's power of co-ordina- 
tion, it would be well to recognize their nature now rather than at some future time. 

Considerable light is thrown upon this question by the techniques which have 
been worked out in the authoritarian states. In every case we find: the abolition of 
representative government; the extinguishment of local government, including home 
rule; the transfer of all policy determination as well as execution to a single executive, 
the legislative assembly becoming a channel for manufacturing consent; the erection of 
an immense bureaucratic machine responsible also for the control of conduct and of 
economic life; the determination of production programs and quotas; the establish- 
ment of a factional party in supreme control and subject to no discipline but its own; 
the rise of a single powerful leader; the disappearance of compromise as a basis of 
action; the stringent control and use of press, radio, schools, universities, public as- 
sembly, and all private associations to condition the whole nation, particularly the 
rising generation; the intensification of separatist nationalism and race consciousness; 
the fabrication of national scapegoats; the stimulation of great mass diversions, par- 
ticularly in sport and expeditions; the extinction of criticism and difference of opinion; 
and the termination of free inquiry. 

It will be observed that there are two great efforts in this whole program, first to 
make the government all powerful in every important sphere of life so that it may have 
the facilities and authority for the co-ordination of all of life; and second, to create 
absolute and universal consent and enthusiastic adherence to the program of the 
state. The attention which is devoted to this latter objective by the authoritarian states 
is a powerful testimonial to the power of ideas in co-ordinating the activities of 

It is clear also from the observation of these experiments, in which direction the 
limitations of co-ordination lie. The difficulties arise from: 

1 . The uncertainty of the future, not only as to natural phenomena like rain and crops, 
but even more as to the behavior of individuals and of peoples; 

2. The lack of knowledge, experience, wisdom and character among leaders and their 
confused and conflicting ideals and objectives; 

3. The lack of administrative skill and technique; 

4. The vast number of variables involved and the incompleteness of human knowl- 
edge, particularly with regard to man and life; 

5. The lack of orderly methods of developing, considering, perfecting and adopting 
new ideas and programs. 

It may also be noted that the authoritarian states are in trouble internationally. 
Some regard this as intentional, as part of the diversion and scapegoat technique, 
while others think that it is more or less inevitable because of the very co-ordination of 
the economic activities of the individual nations in question. If this latter explanation 


is correct, then we may regard the increased friction at the periphery experienced by 
co-ordinated states as a further illustration of the general principle noted above. 22 

It is a striking fact that the authoritarian states have thus far found little difficulty 
in regimenting the thinking and conduct of the masses of the population. This is the 
least of their troubles. The upper and middle classes and the educated are no excep- 
tion to the rule. It is of course too early to accept the dominion of opinion that has been 
established as a long-time phenomenon, and there have been those who have before 
this predicted its early collapse. It must be recorded, however, that there is no indica- 
tion of this breakdown as yet. It is difficult to see how such a collapse can come except 
through a failure of the administration or of the economic policies of the government. 
In other words, the weak link in the chain is not the securing of popular support; it is 
rather in the field of policy and execution. It is at this point that the lack of a system- 
atic method for bringing in new ideas, the corrective of free criticism, and the com- 
mon man's appraisal of the end results may prove disastrous. Certainly a state which 
attempts the extraordinarily difficult task of co-ordinating most of life will need these 
sources of regeneration and correction even more than a state which undertakes a 
more limited responsibility. 

If this analysis is sound, the limits of co-ordination are to be found in lack of 
knowledge and lack of administrative skill. It would therefore be sound policy (1) to 
advance knowledge through research in public affairs, and administration both through 
research and through the adoption of improved techniques. It would seem also 
desirable (2) to advance the area of attempted co-ordination in sectors and experi- 
mentally so that the necessary skill may be gained through trial and error. This would 
mean a deliberate advance in a wave motion, with periods of pause and consolidation 
after each forward move. 

If this course of action is adopted there is no need of accepting the view that there 
are fixed limits of co-ordination beyond which mankind can never go. It is also prob- 
able that the area of co-ordination will always be less extensive than all of life because 
the ability of mankind to co-ordinate, though continually advancing, can never over- 
take the creative capacity of individuals to invent new fields of activity and knowledge. 

The Accretion of Functions 

In view of the fact that organization must conform to the functions performed, 
attention must be given to the process by which new functions are assumed by govern- 
mental units. There have been few studies of functional accretion, and in none of these 
studies has the effort been made to analyze the sequence of events sufficiently to arrive 
at a philosophy of growth. 23 

Certain hypotheses may be stated, however, on the basis of the studies already 
made, such as: 

22 Pp. 21-22. 

23 William Anderson, "American City Government," Henry Holt, New York, 1 925, Chapter XVI, Municipal Func- 
tions; New York State, Joint Legislative Committee on Taxation and Retrenchment, "A Study of the Growth of the 
Functions and Expenditures of the State Government," J. B. Lyon, Albany, 1926 (Legislative document, 1926, no. 28); 
Lent D. Upson, "The Growth of a City Government; an Enumeration of Detroit's Municipal Activities," Detroit Bureau 
of Governmental Research, 1931 ; Sidney Webb and Beatrice Webb, "Statutory Authorities for Special Purposes; with a 
Summary of the Development of Local Government Structure," Longmans, Green, London, 1922, Vol. IV of their 
English Local Government. 


1. Strikingly new activities are generally developed outside of the government by 
private groups and organizations before they are taken over and made a part of 
the normal service of government; 

2. This is particularly the case in highly decentralized governments, and in those 
which do not have a highly developed and professionalized civil service; 

3. In their initiation most governmental services are established on an autonomous 
or semi-autonomous basis, both as to authority and finance and are subsequently 
integrated as to authority and finance with the other activities of government and 
deprived of this preliminary autonomy in spite of their resistance to this change; 

4. Important activities also arise first as special services within a given department, 
and are then made general and set up on an independent footing; 

5. The majority of the services of government have been undertaken under the stress 
of war or of economic depression, and not as advances in periods of prosperity; 

6. Few new services are undertaken after a consideration of their relations to other 
services or of their ultimate costs or benefits; 

7. Private associations, pressure groups, paid secretaries and vigorous new officers of 
civic groups play an important part in bringing government to undertake new 
services. Highly organized communities often "advance their government" more 
rapidly than unorganized communities which may stand in greater need; 

8. A single dominant and original individual either in the community or in the pub- 
lic service may be more influential in the development of new government activi- 
ties than any environmental factor; 

9. The general stream of culture among a people and the universal desire to "keep 
up with the Joneses" are fundamental factors in the adaptation of governmental 

10. Once an activity is taken over by government it is rarely if ever abandoned. 

There have of course been important recessions of the scope of government. 
These should be studied also. The most sweeping of the past three centuries on this 
continent are the separation of church and state and the withdrawal of the state from 
intimate and unquestioned control of economic life. The former was not so much a 
restriction of the functions of the state as of the power of the church. Similarly, the 
latter may be considered not so much a limitation of the function of government as a 
step in the limitation of the power of small entrenched economic groups. When a city 
charter was no longer the charter of a private corporation, and its extensive economic 
powers 24 were passing from "the corporation" to the hands of elected officials, the 
unhorsed economic groups inevitably desired laissezfaire and the separation of business 
and government in much the same way that the church withdrew in the face of the 
advance of the democratic dogma. In both cases there is the appearance of the re- 
traction of the sphere of government. In reality, however, it would seem that the signifi- 
cant element is the change in the "ownership" of the government. But in any case 
these retractions need examination. 

Another type of abandonment of governmental activities takes place with de- 

24 E.g., to grant freedoms, to operate docks and transportation, to regulate all business, to license and control all 
trades and professions, to regulate commerce, and to manage important monopolies, such as "all trade with the Indians." 


mobilization at the end of a war. It will be found, however, that this is primarily a 
process of shrinkage, not of amputation, and that a great many new services are left 
behind when the flood recedes. Here again there are important recent exceptions, as 
when the United States government ceased running the railroads after the World 

In planning for the demobilization of emergency activities, the question arises as 
to whether this may be better worked out when the agency to be liquidated is largely 
independent, or when the agency is a subordinate division of a larger department, the 
remainder of which is permanent. Also, how does the type of organization (i.e. pur- 
pose, process, clientele, place) influence the demobilization? Though little material 
has been collected bearing directly on these questions, it would seem that dependent 
activities, especially those organized on the basis of purpose, are the most readily 
terminated, though it depends, of course, on the nature of the continuing obligation 
involved. After all, demobilization must be carried out by the overhead, and may be 
better done when that overhead itself is not directly involved. 

In view of the growth processes of governmental functions, those who are con- 
cerned with the mechanics of organization will fail to develop a satisfactory theory of 
organization unless they regard their basic problem as dynamic. In considering the 
organization of government we deal not alone with living men, but with an organism 
which has its own life. 

The Evolution of Government 

To what extent is this organism subject to the Darwinian laws of survival? It was a 
basic theory of classical economists that business enterprises survive only in so far as 
they adapt themselves to the changing economic environment, and that only the fittest 
survive, fitness being measured in terms of prices and profits. Whatever the truth of 
that hypothesis, mankind has determined not to try it out. At one end business has 
upset the free competitive test by monopolies and cartels, and at the other end the 
public has refused to let itself be pushed around freely by such combinations or to let 
business enterprises go to the wall in any wholesale fashion, when a whole system can- 
not meet the economic judgment day. 

When we turn to governmental organizations we find even less "survival of the 
fittest." Governmental organizations seem to be extraordinarily immune to evolu- 
tionary changes. Next to the church, they are in all civilizations the most vigorous 
embodiments of immortality. 25 A governmental unit is by nature a monopoly, and is 
thus not subject to the purifying influence of competition. It does not have a profit 
and loss record; its balance sheet is buoyed up by "good will"; its product is priceless 
and often imponderable; its deficits are met from taxes, loans and hope. Under these 
conditions a governmental unit can continue for many years after its utility has 
passed, or its form of organization or program have become obsolete. 

25 The church has, of course, great advantage over the state in that the church deals with the inner spiritual life, and 
does not often come into direct physical conflict with those embattled forces which arise in the world of iron, coal, oil, 
labor, boundaries, colonies, wages, land, security, and airplanes. When the church does speak on these topics, it generally 
leaves the battle to other organizations, principally to the state. Under such conditions it is the state and not the church 
which runs the risk of extinction. A good illustration is the comparative mortality of the church and the state south of the 
Mason-Dixon line in 1865. 


Mankind is extraordinarily docile and tolerant. Though men are born with the 
"unalienable Right" 

"to alter or abolish their Government and to institute new Government, laying its founda- 
tion on such principles and organizing its powers in such form, as to them shall seem most 
likely to effect their Safety and Happiness ... all experience hath shown that mankind 
are more disposed to suffer, while evils are sufferable, than to right themselves by abolish- 
ing the forms to which they are accustomed." 

While the immortality of governmental institutions thus reflects the natural con- 
servatism of mankind and the tremendous force of inertia in human institutions, it 
shows also the ultimate elasticity of governmental institutions. Governments generally 
mend their ways and their policies so that they may survive. This is particularly true 
in a democracy. A democracy is characterized by the fact that there is built into the 
structure of the government a systematic method of introducing changes in program 
and method as the result of the broad movements of public opinion. As a result demo- 
cratic constitutions should be more elastic, more subject to evolution, and therefore 
more immortal than other constitutions. 

The struggle for survival in government thus becomes not so much a fight to the 
death, a test to destruction, but an endless process of adaptation to changed conditions 
and ideas. In this sense, governmental institutions are in continual evolution. But the 
process of evolution of human institutions is quite different from the process of evolu- 
tion of living organisms. 

The process of adaptation falls partly in the field of politics and partly in 
the field of administration. The two are so closely related, however, that the 
political aspects cannot be ignored completely even here, where we are con- 
cerned only with administrative organization. A glance at the present world 
situation makes it clear that the modern state faces as never before the need of 
rapid and radical adaptation to changed conditions. Governments which cannot 
make the necessary evolutionary changes will not survive. It becomes necessary, 
therefore, in the structure of the organization to make more elaborate provision for 
those agencies of management which concern themselves with the processes of adap- 

What are these agencies? Are they not: 

First, those which have to do with political life and leadership; 

Second, those which have to do with getting new work promptly undertaken and 
efficiently done; and 

Third, those which have to do with the careful observation of results? 

This is not the place to discuss the first item, politics. From the standpoint of 
organization, however, it does seem that insufficient attention has been given to 
political life and leadership. Except in England, with its long tradition of collegiate 
executive responsibility, the political life of the nations of the world seems to be drifting 
into the hands of a single strong leader, and the currents of political life to be more and 
more confined by dikes within a narrow channel. While this may be the short cut to 
effective and immediate adaptation of the program and organization of the state, 
human experience seems to indicate that this structure of political life will have pre- 
cisely the opposite result in the long run. If there is any truth in this observation. 


it would seem of extraordinary importance to take steps to expand the structure of 
political leadership and free the currents of political life. There must be leadership with 
increased scope for action, but every step toward increased power must be matched 
with a step toward increased accountability. 

Instead of superseding or destroying legislative bodies, consultative institutions, 
and independent examination and audit because they stand in the way of quick 
changes in governmental programs, has not the time come to strengthen these organi- 
zations especially as agencies for the effective examination and criticism of policies 
which have been put into operation and thus as agencies for the more orderly consider- 
ation of new policies for the future? 

In the field of organization the needs are clear. In periods of change, government 
must strengthen those agencies which deal with administrative management, that is, 
with co-ordination, with planning, with personnel, with fiscal control, and with re- 
search. These services constitute the brain and will of any enterprise. It is they that 
need development when we pass from a regime of habit to one demanding new think- 
ing and new acting. 




L. URWICK, O.B.E., M.C., M.A. 

Chairman, Urwick Orr and Partners, Ltd. 
Consulting Industrial Engineers 


Director of the International Management Institute 




1 . Need for a Technique of Organization 

It is the general thesis of this paper that there are principles which can be arrived 
at inductively from the study of human experience of organization, which should 
govern arrangements for human association of any kind. These principles can be 
studied as a technical question, irrespective of the purpose of the enterprise, the per- 
sonnel composing it, or any constitutional, political or social theory underlying its 
creation. They are concerned with the method of subdividing and allocating to in- 
dividuals all the various activities, duties and responsibilities essential to the purpose 
contemplated, the correlation of these activities and the continuous control of the work 
of individuals so as to secure the most economical and the most effective realization of 
the purpose. 

In existing world conditions, the practical importance of this subject can scarcely 
be exaggerated. In every aspect of its common life humanity is registering failures and 
enormously costly failures in its capacity for purposeful association. The majority of 
all nations earnestly desire peace: the machinery of peace works haltingly and ineffec- 
tively. In country after country liberty of speech and of person are lost, because demo- 
cratic institutions fail to devise an administrative structure adapted to the speed and 
complexity of social evolution. The world's productive equipment is manifestly capa- 
ble of yielding vastly increased quantities of goods and services: millions starve because 
the financing and distribution of this plenty are not organized. A machine technology 
points to the obvious economies of large-scale units of business control: amalgamations 
founder because there is widespread ignorance as to the methods of managing these 


In 1931, under the title "Onward Industry," 1 Messrs. James D. Mooney and 
Alan C. Reiley published a full-length book examining the comparative principles of 
organization as displayed historically in governmental, ecclesiastical, military and 
business structures. The general outline of their concepts is shown in Figure I. Their 
book constitutes the first serious attempt to deal with the subject comparatively and 

For the purpose of this paper it is sufficient to note their insistence on the impor- 
tance of co-ordination: 

"This term expresses the principles of organization in toto; nothing less. This does not mean 
that there are no subordinated principles; it simply means that all the others are contained 
in this one of co-ordination. The others are simply the principles through which co-ordina- 
tion operates, and thus becomes effective." 2 

Co-ordination operates in two different senses. If the total of the activities in- 
volved in any enterprise are represented as a square plane, the various groups of activ- 

1 James D. Mooney and Alan C. Reiley, "Onward Industry." Harper & Brothers, New York and London, 1931. 

2 Ibid., p. 19. 






















OK (0 O 03 O -P -H 

J 1ilia!ll 

1 5- J*8P P* T 


flit! 1 !! 


sir J 111 
\ ss s •ill* 

i a S .a5a a -Srt". 


*3 3 

o o ^ *3 +> g 

3 H)-H ffl q 
<i «h o a "O to 



C O & D o 

^ an BtI 
I I 

"if 8 ?S! 

c -a h o a a 

■ +> d o o 

:i»» i-a 

c o a 3 

II alas 


1 • 6-ao ■ 

I "IPs 

Js E 




O Q ,5 P - P *3 


oh o ■** o 3 


I if | 

B V P 

■h g J> 
n G +> 

3 fit 8 

+> .a .e b ^>+* n i-« 

<4> O « O 

■P+"fl o c 
■h . E o o 5 

u. c ci c a u - 

O O^fl B£ C' 

13 *J U *3 -H 

H <t| Pi n E • 

9C 60 d a «r 
, O P rj *■ CO *. 

K ■♦* W 5 C & •* 

S flj G C 01 '■■. 

fc H B -h -h c l 
§ "3 n "5 8 £ ' 

w o & M * j 

i sf sss . : 

•fl o jjj « p o i 


d t ov c« 

Of O +• 5) O ■ 

P< C C O E I 

« « -P B U i 

>,-H )) D) D d 

e £ -h x o pi 

P C • -H h 

a 2 - +> 

Ct) M • o c. 

O 1> U • c o 

■HMO ■ 3 -H 

+* T( • ti 4-> ' 

O C « C s - 

B WE- -H o u - 

6 ■ O X E : 

4J « S e c h 


•H 4* 3 • 


H B 8- 

s 3' 

H OS « 

I fill 

U O (H « 

q -a <3 o 

So -H 
o *j 

' a B 3 S S 


i p "3 1 
















ities will be found to be divided both by vertical and by horizontal lines drawn up and 
down and across the square. The work of individuals is always divided by vertical 
lines into different tasks. Such subdivision is usually either "serial" — the responsibil- 
ities follow each other in process, "unitary" — the responsibilities are defined by 
areas or objects, or "functional" — the responsibilities are distinguished by kinds or 
subjects. But where any large numbers are concerned, work must also be divided up 
by horizontal lines into different levels of authority and responsibility. This grouping into 
levels is the special thesis of this paper. The adjustment of the resulting authorities and 
responsibilities to each other and their continuous correlation constitute one of the 
main tasks of leadership in any enterprise. 

3. The Scalar Process 

The evolution of ideas on this question has been comparatively simple. Originally 
almost all undertakings were organized on what has been called the "line" or, incor- 
rectly, the "military," principle. Emphasis was placed on the importance of what 
Mooney and Reiley have described as the "scalar process" 3 — 

"The supreme co-ordinating authority must rest somewhere and in some form in every or- 
ganization. ... It is equally essential to the very idea and concept of organization that 
there must be a process, formal in character, through which this co-ordinating authority 
operates from the top throughout the entire structure of the organized body." 

The considerations which appeared of greatest importance were that there should 
be clear lines of authority running from the top into every corner of the undertaking 
and that the responsibility of subordinates exercising delegated authority should be 
precisely defined. Since, in all cases, concrete objects, physical boundaries or the 
limits of some well-known technical process, offer the simplest and readiest means of 
definition, the unitary or serial methods were almost universally adopted in sub- 
dividing and grouping activities into tasks. Division into levels followed the method, 
A supervises B-C-D and E; B supervises P-Q-R and S; P supervises W-X-Y and Z 
and so on, the supervisor at each level being totally responsible for every aspect of his 
subordinates' work. 

4. Specialization 

Modifications of this arrangement have occurred throughout human history. 
Wherever human knowledge and skill have been specialized, for instance in the case 
of the law, somehow and somewhere that specialized authority has had to be inte- 
grated with overall authority and responsibility. But, since the introduction of power- 
driven machinery and the great advances in applied scientific knowledge of the nine- 
teenth century, the amount and complexity of specialized skill required in connection 
with every kind of human enterprise have vastly increased. 

The first formal recognition of the consequence of this tendency for previous ideas 
about organization came from F. W. Taylor. Imbued as he was with the necessity of 
basing industrial management upon exact measurement and specialized knowledge, 
he was impressed with the impossibility of discovering subordinates who could exer- 
cise overall responsibility with maximum effectiveness in respect of all its aspects. In 

3 James D. Mooney and Alan C Reiley, op. cit. 


this connection he developed two of his well-known principles of management. The 
first was the separation of planning from performance — 

"All possible brain work should be removed from the shop and centered in the planning 

The second was the substitution for the older type of organization of what he called 
"functional" management. 

"Functional Management consists in so dividing the work of management that each man 
from the assistant superintendent down shall have as few functions as possible to perform." 4 

By a "function" Taylor meant a particular kind of work, a subject. 

Under these principles as Taylor operated them each worker was managed by 
four foremen in the shop specialized in certain aspects of his task. The task itself was 
set by four further specialists working in the planning department. This part of his 
general management practice has been applied less, and with less success where at- 
tempted, than any other feature of his work. It must be recognized, however, that his 
ideas on the subject rendered an immense service. In the first place the removal of 
much of the routine of management from detailed personal control to the operation of 
system eliminated an important source of friction from industrial life. In the second 
place his insistence on the need for specializing the work allotted to individuals in or- 
der to command exact and scientific knowledge made possible immense advances in 
the art of management. On the other hand his mis-description of the older type of 
management as "military," has discouraged students of organization from turning to 
the one field where the need for co-ordination is overwhelming and in which expe- 
rience in the control of large numbers is centuries old. Taylor's experiments were 
chiefly concerned with a single department — usually a machine shop. And even in 
such cases co-ordination was secured through a Superintendent in charge of the shop 
as a whole, and an Assistant Superintendent who looked after the Planning Room. 
But, when his conception of a division of responsibilities by function is carried higher 
up the line of control in any large enterprise difficulty is encountered immediately. 
The "scalar process" is weakened. Co-ordination lapses or is secured only by excep- 
tional efforts. 

"When a considerable amount of staff organization is introduced . . . this tends naturally 
to weaken the disciplinary effects of line control; and where staff organization is used to any 
marked degree special care must be used to supply co-ordinative influences to compen- 
sate for this weakness." 5 

5. The Span of Control 

For this, there is a quite definite reason. Students of administration have long 
recognized that, in practice, no human brain should attempt to supervise directly 
more than five, or at the most, six other individuals whose work is interrelated. Mr. 
A. V. Graicunas of Paris has shown why this is so. An individual who is co-ordinating 
the work of others whose duties interconnect must take into account in his decisions, 
not only the reactions of each person concerned as an individual, but also his reactions 

* Cf. F. W. Taylor, "Shop Management," pp. 98 and 99. 

6 Dexter S. Kimball, "Principles of Industrial Organization," p. 63. "Staff" as used in this quotation implies a 
man or department specialized in a particular function. 

j» ei ^ 

— n to 
— as 

- = g 

2 g 2 

ck « tQ 

a) *c eo 

r^ e> — 














g s 

2 § 

P «» 

•< _ 

-J C 

a 3 

C .6 

in a 

en w 

o < 

a § 

g i 

< » 

o 1 

u Cj 


o : 




3 ■» 


e _ 

I « 






















S3 S 

C^cN Iff* 

c e 







c |c< 




e slot 



s 7« 






.1 II 




























I iil 

3.= L. O C c 
.a u — — n 

3 S- 




g a? 




3 "d 


oj a - 
a u ^5a 

b B"„ 


|-3 J § 1 

- « O 3 

ir°i J 

U U X — 

CB J3 * -C -. 

K u " o a 

« = = a £ 

v D C — 

« "9 5 - M 

— .- OJ c fc. 

. s . >> 

S O £ fc 

•5 = S X 

s I § a 

1 o I * 

V « V 

as j, o= 

= 2 

e = 


5 * a' 

.«• « 

S Si 

Q; -C ^ 

■H O « o 

q, « £ a 
•> =5 g 

S u *> o 

2 " S * 



. o 




1 e 

a s 

• = 


i a a a 
r o a S g 

a uS 



l ffl q o 
" o ffl 
i o u, 

03 [d Id 
1 QQ 


o < a 


t a>< 

a S 


u <g 

< Q 


< < 

a ug gin 
o°<o say 

bK <u goo 

fc- rl fie. O 



ou - a. b < 

J a Q o * fl r-i 

5 o u,S° u 

(- n oop°oo 
M o„« o2 n 9 

J3 OO o te. < S 

5 yao° fc ou 
c a g q s s o o 

,aio u < o 
aa > &[<.<o o2 


•- ■" 5 


7i •= ■* 5 

S S 'S "a 

i a q 

% 2 5 2 5 | 

4 H i h 

O cc 
£ o _ = 
•r 8n o ■= 

X til 


■a .a ■a Z ■• .. + 
E E E si ° * 

3 3 3 5 + + + 

= « = b ■ i ; 

1 1 1 1 1 1 1 


° i a 


o e ™ 

■- E = 






u o 


a< a S 2<u 


a. I 

c E 

< S2 fc 
b n< g 


<2u o 

n ,« «i s 

feu' 5 

3- & 


- s 



















as a member of any possible grouping of persons which may arise during the course of 
the work. 

The psychological conception of "the span of attention" places strict limits on the 
number of separate factors which the human mind can grasp simultaneously. It has 
its administrative counterpart in what may be described as "the span of control." A 
supervisor with five subordinates reporting directly to him, who adds a sixth, increases 
his available human resources by 20 per cent. But he adds approximately 100 per cent 
to the complexity and difficulty of his task of co-ordination. The number of relation- 
ships which he must consider increases not by arithmetical but by geometrical pro- 
gression. The operation of this principle is shown graphically in Figure II. 6 

Neglect of the limitations imposed by "the span of control" creates insoluble 
problems in co-ordination. Two examples are given in Figures III and IV. 

The first shows the organization of the Secretariat of the League of Nations pro- 
posed by a Committee which reported in 1930. Arrangements are made for no less 
than 15 independent officials to report direct to the Secretary-General. When it is 
remembered that, in addition to this administrative task, he has also to maintain 
personally, political and diplomatic relations with the individuals composing the 
heterogeneous and constantly changing delegations of more than fifty states, members 
of the League, some of the breakdowns in the machinery of peace, call for little further 
explanation. The organization is totally inadequate to the complexity of the task. 

The second shows in outline the administrative organization of the British Gov- 
ernment. In the Cabinet alone the Prime Minister has to co-ordinate the work of 17 
ministers. There are as many or more departments not represented in the Cabinet. 
The provision of a number of separate departments dealing with functional questions 
in relation to Scotland only is an interesting example of the persistence of local senti- 
ment and traditional arrangements, despite altered circumstances and a great increase 
in the complexity of government. Such mechanisms cannot apply the scientific knowl- 
edge which is available to the task of administration. 

Specialization is essential. Without it men cannot be found capable of placing at 
the service of any undertaking the most recent and approved methods in relation to 
each of its activities. On the other hand, if it is to be effective, specialization must be 
to some degree authoritative. This consideration inevitably complicates the work of 
co-ordination. In a business enterprise, for instance, a chief executive with a pure 
departmental organization who wishes to market a new product, X, will simply tell 
the manager of the department responsible for that class of product to go and make 
X and sell it. But a chief executive with a highly specialized organization must bring 
all kinds of officials together to plan out the part that each has to play in producing 
and selling X. Plans having been made, he must supervise each separate specialist to 
secure that his work coincides in time, space, quantity and quality with the work of 
others. Moreover, conflicts between specialists of different kinds or between specialists 
and general managers are difficult to avoid. 

"The great difficulty lies in getting the members of the Staff Departments and those of the 
Line Departments to co-operate." 7 

6 For a fuller exposition of Mr. Graicunas' theory, vide his paper in this collection, entitled "Relationship in Organiza- 
tion," pp. 181-187. 

7 Paul M. Atkins, "Factory Management," p. 39. Here again "staff" is used of a department specialized in a 
particular function. 


o e 

e u 

•H • II 


3 -< «• 
a o 



tf * © 

V « «• 
O • Q 

** o v. • 

« T3 4* © 

i3- 3 -j 

4» C3 -9 • ** 

a -^ -i js o 

• 5 g a o 

h 9 d a 

s - 1 -a s « 


-< H © « 

a> tf <? 

3 4» ♦» o «• 

O B 
•W -O *a 

■g g .= -2 8 g 

O • •** C I* © 


3 « « J& <D J* 

5 t 


v • « ** o d 

V f-T B r-i 

« « -H - W 

j3 qq <d m en 

8 fiSB"i 

4-t C ■) © a-l 


o « a e n 



ft u u is u 

• O l> 1 «* 



t4 4* « *] « 

§ * +» © 


L> fJ O O 


(US © ■*> 
*» B c j3 

C O * *» H 


coo o 

o f> u u 





Co © 


« -t c 

Jt CO 


O Vi 

u o 



4 1 

S 5 . 


«H €> 


4» 4* 


o ♦* 

» ct» 

p-i U » 

V 4* 

© *H 

4* um. 

*a a. M 

« n 


£ «fl W 

r4 •* 

e • • i ■ 

^ • " Vh c ** 

4* © C YHttV ** «■ 
IriH VII 4 «f* 


TT - 









■P I- 



-4 e 

►» o 






*H • 

a • 

m n 










■Lord Chancellor" 
•Home Office- 


ral Oueetion of Foreign 

ory Legislation & itB 

venue & Commercial Oueetion 
S. TariffB 

Compulsory rationalization 
of an industry 

ployment relief by public 

tion for administration 






PoBt Office- 

Board of inland- 
Auditor General- 

Lo^al Government- 


art is 
ably in 
chan ism 
on ,mos t 
To get 
-ailed proced 
one subject n 
research .But 
great complex 
-i nation & th 
any logical g 
functions lix 
a .successful 

in<fomple te 
accurate 1 1» 
are many otk 
s of coord- 
ly committee 
all the det- 
ure on any •" 
eeds much 
it shows th© 
ity of coord 
e absence of 
rouping of 
ely to yield 



In every undertaking the "scalar principle" must be observed or authority breaks 
down. If the subdivision of the authority of the leading officials is on "functional" 
lines, this subdivision recurs throughout the enterprise. Decisions multiply which call 
for a correlation of different "functional" views, a correlation which can only be au- 
thoritative if made by the chief executive himself. At the same time there must be 
subordinate co-ordinating points. A planning room in New York cannot manage a 
factory in San Francisco. But when this is admitted the chief executive finds himself 
with immediate subordinates both on a "functional" and on a "unitary" basis. There 
is a tendency for their number to exceed his "span of control." If he groups "func- 
tions" the same difficulty of cross-correlation occurs at lower levels of the organization. 
The dilemma is a real one. In a large organization the complex of different principles 
which demand consideration in the structure of authority and responsibility may be 
most serious. 

An example of the problem is offered by the British postal organization. Under 
the Postmaster General, a political appointment, the Director of the Post Office is the 
chief of a very large staff. Obviously some form of area delegation of the work of 
co-ordinating Post Office services and activities is indicated. On the other hand, the 
work calls for specialized supervision in certain important respects, notably finance, 
engineering and personnel. But in the minds of the public who control the Post Office 
through Parliament, it is the various different services — mails, telegraphs and tele- 
phones — which are the focus of attention. Parliamentary questions are apt to take 
the form of "Who is responsible for the telephones?" 

There are thus three different requirements which should be met, and must all 
to some degree be met, in delegating authority — area co-ordination, control of spe- 
cial functions and responsibility for different services. How is the Director of the Post 
Office to secure that his work is properly supervised and co-ordinated in each of these 
respects, without infringing the principle of "the span of control"? Yet that principle 
is absolute. In so far as the chief of any enterprise departs from it, he renders his con- 
trol weaker and less effective. 

6. "Line and Staff" 

The solution so far adopted in practice is known as the "Line and Staff" system 
of organization. It is admittedly a compromise. 

"The line and staff type of organization is a combination of the best features of both the line 
and functional types" 8 

or again 

"Line and staff organization combines features of both the line and functional types." 9 

Precise statement of what is meant by the term is lacking. A form of organization 
may be described with reference to the functions allotted to the various positions or 
with reference to the relations between different positions. 

On the first point writers on business administration give divergent opinions. 
Thus "staff" positions or departments have been variously described as those which 

8 N. B. Cornell, "Industrial Organization and Management.' • L. P. Alford, "Management's Handbook." 


a. "deal with one particular phase of business." 10 

b. "give expert advice to the line officers. The primary functional divisions are Sales, Engi- 
neering, Manufacturing and Finance." " 

c. "are in charge of a single staff function or certain similar or supplementary functions." n 

d. are engaged in "the business of analyzing, testing, comparing, recording, making re- 
searches, co-ordinating information and advising." I3 

e. "furnish the means of developing standards and plans for which a regular line position 
leaves no time." H 

f. "give special types of services." 15 

g. are "responsible for investigations, study and designing. . . . Routine executive direc- 
tion of performance usually requires the rapid shifting of attention from one immediate 
problem to another, leaving little time either for reverie or for continuous concentrated 
attention to a single idea." 16 

The dominating idea is that of specialization — Taylor's conception. Any function 
which is specialized may apparently become a "staff" function. 

On the second point there is a greater measure of agreement. The majority of 
writers emphasize that the relation of the staff to other departments is advisory. 

a. "The essential feature of the staff organization is that it is purely consultative and ad- 
visory." 17 

b. "Fundamentally the staff departments are advisory in their character." 18 

c. "The staff man usually serves in an advisory capacity to a line executive. He frequently 
has no authority." 19 

But on the other hand 

d. "The line supervisor ... is responsible for seeing that the recommendations and in- 
structions of the staff experts are carried out." 20 

e. "In theory, a staff department has complete authority in its particular field in so far as 
policies and methods are concerned." 21 

In a few business concerns which have paid close attention to organization the 
relationships between "staff" and line officials have been defined closely. Messrs. 
Mooney and Reiley have perceived that the "staff" function has not only informa- 
tional and advisory phases, but a supervisory phase which includes the services of 
inspection. As might be expected, the General Motors Export Company of which Mr. 
Mooney is President has definite and acceptable ideas on this point. 

"A line officer exercises authority over all of the body of organization lying beneath him on 
the chart, whereas the influence exerted by a staff officer outside of his immediate depart- 
ment is, so far as it is authoritative, an authority of ideas. The staff officers are, in their 
functional capacities, responsible advisers to their respective line superiors, and advisers 
also to the corresponding staff officers in the subordinate organization strata, but any direct 
line instructions they may wish to see promulgated may be promulgated only back through 

10 B. H. Lansburgh, "Industrial Management." 

11 Dexter S. Kimball. "Principles of Industrial Organization." 

12 W. B. Cornell, "Industrial Organization and Management." 

13 H. S. Dennison, "Organization Engineering." " O. Sheldon, "The Philosophy of Management." 
15 Paul M. Atkins, "Factory Management." 16 H. P. Dutton, "Principles of Organization." 

17 O. Sheldon, op. cit. 18 Paul M. Atkins, op. cit. 10 H. P. Dutton, op. cit. 20 W. B. Cornell, op. cit. 

21 R. C. Davis, "Factory Organization and Management." 


their line of contact with their superiors and down thence to the line officers in the next sub- 
ordinate stratum. ... If the expression descriptive of a staff officer's authority — an 
'authority of ideas' — means anything at all, it means that the staff executives' plans and 
recommendations are entitled to the respect and consideration of the line executive. A very 
definite burden is, therefore, put upon the line executive who sees fit to disregard or to reject 
the counsel and help of his staff associates. . . . Four cardinal principles . . . enter into 
a proper understanding of the relationship between line and staff. . . . 

1 . Line and staff are jointly responsible for performance. 

2. A staff officer discharges his responsibility by furnishing information and advice 
which he makes available to the line officer unselfishly and without thought of 
personal credit for the results accomplished. 

3. Although staff executives are charged with responsibilities that have to do with 
internal administrative phases of the work in their own departments, this does not 
give them direct authority over the line forces in subordinate organization strata, nor 
does it relieve their line superiors of the basic responsibility for the results of their 

4. The line recognizes the purpose and value of the staff and makes full use of its 
advice and assistance. In order that the line may properly do so, the staff must 
create for itself an authority of ideas, and must, by competence and tact, obtain and 
justify the line's confidence." n 

7. Emphasis on Relationship 

Indeed, hitherto, the major emphasis in business organization has been placed 
on this question of relationships. There is a natural resistance among subordinates in 
all undertakings to any new form of specialization. Officials, as human beings, resent 
and obstruct any suggestion that functions which they are performing and authority 
which they are exercising should be transferred to another. An historic and very amusing 
example is found in Lytton Strachey's account of the difficulties raised by doctors of the 
Royal Army Medical Corps when Florence Nightingale first suggested the specialization, 
and, last horror, the feminization, of the function of nursing. When that very determined 
lady insisted on unpacking shirts for wounded men who were freezing, before a Board of 
Audit had sat on these supplies — a procedure which usually occupied three weeks — 
the Purveyor stood helplessly by "wringing his hands in departmental agony." 23 

Leaders are apprehensive of any weakening among subordinates of the sense of 
full responsibility and hence of discipline. They recognize that if responsibility is not 
direct, the "scalar process" will break down. As the work of subordinates becomes 
more specialized it is apparent that, in the absence of any adequate solution of the 
problem of co-ordination, the overall control of the chief will be less effective. 

In consequence almost every discussion and every experiment concerned with 
the introduction of new specialist supervision, centers round the relationships with the 
existing line officials, the safeguarding of their authority and the protection of the 
direct chain of responsibility. Business has become more concerned in determining 
what specialized officials are not to do, than in studying what they should do. It does 
not define a man's functions and responsibilities in an organization to tell him that 
they are "purely consultative and advisory." It merely throws upon him the respon- 
sibility of securing executive action, without giving him the necessary authority. 

22 Cf. "Organization and Operating Principles," E. W. Smith, Handbook of Business Administration, pp. 1474-1488. 

23 Cf. Lytton Strachey. "Eminent Victorians." 


Inevitably there has been tacit evasion of the more important question as to 
what should be the duties and functions of officials in a "staff" position. This has 
added to the confusion of thought on the whole subject. There is reason to suspect 
that the compromise does not integrate the two principles. The necessity for special- 
ized authority is accepted in theory. But the difficulties which it raises in practice in 
relation to co-ordination are concealed in verbiage. It is said that the specialists have 
no authority or only "an authority of ideas." In effect such phrases imply a surrender 
of specialized control. 

8. Effect on Co-ordination 

More important, if the various functions and relationships assigned to "staff" 
officials in books on industrial management are a correct description of practice, they 
carry the implication that such officials cannot, by the very nature of their work and 
position relieve their chiefs of any of the increased burden of co-ordination which is 
the central theme of this paper. It is impossible to contend that an official is diminish- 
ing his chief's co-ordinating task: 

1. If he is specializing in the work of some function previously performed by 
subordinates in the normal "line" organization. He is merely adding one to the 
number of individuals whose views the chief has to consider, and the geometric pro- 
gression of the number of direct subordinates whom his work affects, to the relation- 
ships his chief must take into consideration. 

2. If he is a specialist himself in some particular function. The main task of co- 
ordination on the intellectual side is to synthesize the views of various specialists and 
to iron out the bias which specialization inevitably produces. If an official is himself a 
specialist he merely adds to the number of opinions which his chief must analyze 
and correlate in his own mind. Moreover, co-ordination is essentially a question 
of relationships. As has been pointed out, "specialization tends so to narrow the 
executive's appreciation and knowledge of relationships as to make it impossible for 
him effectively to create and to maintain those relationships for which he is re- 

3. If he combines staff and "line" functions, other than the immediate control 
of such few subordinates as are necessary for investigation and record in connection 
with his own proper work. If he is responsible for any of the main functions of the 
undertaking, an executive bias is added to the intellectual bias mentioned in 2. In 
addition this form of arrangement violates both Taylor's main principles. The official 
has not "as few functions as possible to perform." His position does not segregate 
"planning from performance," a feature emphasized by a number of writers as the 
fundamental characteristic of the "staff" concept. 

4. If his relationship to his chief or to this chief's principal subordinates, is 
"purely advisory." Co-ordination is essentially an executive function. No part of the 
executive burden of co-ordination is removed from a chief's shoulders by the receipt 
of advice, additional to that which he should seek in any event from his principal sub- 
ordinates. The chief purpose of co-ordination is to secure detailed correlated action by 
individuals: the chief obstacles it has to overcome are differences of outlook or empha- 
sis leading to heterogeneous initiatives. An official who is only entided to "give ad- 


vice" which may or may not be accepted, cannot relieve his chief of any part of the 
personal difficulties involved. 

5. If his main work is investigational. A "staff" official may well undertake in- 
vestigations for his chief. But, unless those investigations are synthetic, are designed to 
bring together various functional aspects of a particular problem, they do not con- 
tribute to the task of co-ordination. If they are concerned with a single "function" or 
subject, they might equally well be undertaken by an official in any other relation- 
ship. Their assignment to a member of the "staff" merely indicates that the organiza- 
tion is not equipped to deal with that particular function or subject. 

"A staff is something to lean on." 24 A chief cannot lean on a "staff" official who is 
adding to, rather than decreasing, the burden of his main preoccupation. 

9. The Problem of Terminology 

One reason for this situation appears to be a lack of discrimination in terminology. 
The "staff" conception is admittedly borrowed from military life. But in military 
phraseology the term "staff" is used by a number of armies in two senses. The wider 
sense implies all specialized troops and services; the narrower sense, sometimes quali- 
fied as "general staff," implies selected officers who assist the commander in carrying 
out his functions of command. These two groups, the specialist troops and services 
and the "staff" proper, have completely different functions and relationships in army 
organization. Even the most authoritative students of business organization appear to 
have missed the full significance of this distinction. 

The terms used in the United States army distinguish between "general staff 
officers" and "technical and administrative staff officers." It is interesting that so 
sympathetic an authority as Mr. H. S. Dennison actually quotes from a United 
States army manual which makes this distinction without, apparently, appreciating 
its significance from the standpoint of business practice. 

"General staff officers assist the commander by performing such duties pertaining to the junction 
of command as may be delegated to them by regulations or given them by the commander. 
Technical and administrative staff officers assist the commander and his general staff in an advisory 
capacity in matters pertaining to their special branches." 25 

Mr. Dennison comments: 

"Business uses the term 'functionalizing' to distinguish, though not always very clearly, 
the establishment of departments to give special types of service, usually advisory, to the 
more strictly operating departments. They correspond to what in the older Army parlance 
are called 'staff' departments or 'staff' officers in distinction from the 'line.' . . . The his- 
tory of the evolution of army organization as it has worked itself out over many centuries 
is rich in suggestive material." 26 

Mooney and Reiley note that: 

"The modern military application concreted in the term 'general staff' is something to 
which the student of organization must give his careful attention. . . . There is a vital 
respect in which military staff organization is in advance of anything yet developed in the 

24 Edgar W. Smith, op. cit. 

25 "Command, Staff and Tactics," General Service Schools, Fort Leavenworth, quoted by Dennison. Italics are the 
author's. 26 H. S. Dennison, "Organization Engineering," pp. 144-7. 


average industrial establishment. This is in the service of transmission of line decisions to 
everyone concerned. . . . Although co-ordinated staff service in a form that resembles the 
military general staff is growing in business institutions, the office of Chief of Staff or any 
office under any name existing solely for the purpose of co-ordinating these functions is 
unknown in industrial organizations." :7 

Other writers who discuss methods of co-ordination do not mention "staff" as 
a means to this end. 

"Of the several mechanisms that are in use to secure co-ordination of effort and executive 
control the most important, perhaps, are organization charts, organization records, stand- 
ard procedure instructions, orders and returns, records of performance, administrative 
reports and committees." !8 

In the British army the distinction between staff and specialized functions is 
definite. Those who "assist the commander in the execution of his functions of com- 
mand" are known as "staff officers" irrespective of their special branch: "General 
Staff" is used to describe the particular branch of staff duties concerned with opera- 
tions as contrasted with the branches concerned with administration. The troops 
commanded and administered by the commander and his staff are divided into 
fighting troops and services. "Fighting troops carry out the actual operations." Their 
maintenance is secured by "the services" which "provide them with all their require- 
ments in personnel, animals and military material." 29 

Both these groups are themselves considerably specialized — the fighting troops 
into various technical arms and the services into various administrative functions, 
supply and transport, equipment, medical, veterinary and so on. In practice the 
"line" really consists of Infantry and Artillery — the principal arms. The "staff" 
co-ordinates all arms and services. But the technical fighting troops — engineers, 
signals, machine guns etc. — are co-ordinated with the "line" almost exclusively 
through the staff, and the services report to the staff in accordance with regulations. 
The Commanders of Infantry and Artillery formations in a division are senior to, the 
commanders of units of the more important technical troops and services are the 
equals in rank of, the principal staff officers. 

The duties of "the staff" are laid down with great precision. 

"It must be the main object of staff organization to ensure a smooth and efficient co-ordi- 
nation of effort between all portions of the force." 29 

There is a very wide distinction in character between the command of technical troops 
or specialized services, i.e. in business parlance, the control of specialized staff depart- 
ments, and the co-ordinative duties allocated to the "staff." The principal officers of 
such technical troops or specialized services carry out some "staff" duties in that they 
act as advisers to the commander on the subjects falling within their field. But nine 
times out of ten he will seek such advice through the appropriate staff officer. The 
staff officer's relations to such specialized or technical troops and services are closely 

27 Op. cit. 2S Cf. Dexter S. Kimball, "Principles of Industrial Organization," p. 103. 

29 Quotations from British "Field Service Regulations," Vol. I, Organization and Administration. 


"The commander's staff is responsible that his intentions are known and understood . . . 
is charged with the issue, both to the services at headquarters and to the subordinate com- 
manders, of such orders and instructions as will enable co-ordinated action to be taken. 
. . . Staff officers alone have authority to sign on behalf of commanders. Every order 
and instruction issued through the staff is given by the authority of the commander and on 
his responsibility." 


"the commander and his principal staff officers are responsible for decisions on technical 
and financial questions only when the head of a service refers such questions to them and 
. . . the power of interposing in questions of this nature is confined to the commander." 29 

The Staff Officer is not himself a specialist. 

"The heads and representatives of services are the advisers of the staff in regard to matters 
connected with their services, and the staff should consult them before forming an opin- 
ion. . . . The methods to be adopted should be left to the head of a service to determine." 


"while it is necessary that the work of the staff should be distributed in accordance with 
the nature of the duties to be performed, . . . there is but one staff, having but one pur- 
pose. ... It is essential for efficient staff work that officers of each branch should have a 
practical knowledge of the work of other branches. . . . The relationship between all 
officers serving on the staff must be close and cordial." 29 

The Staff Officer has virtually no "line" functions, other than the control of such 
subordinate staff officers and clerical staff as are required for his own special work of 
co-ordination. His and their duties are: 

a. The collection of information for the assistance of the commander and its dissemina- 
tion to his subordinates. 

b. The transmission of the commander's orders and instructions. 

c. The anticipation of the difficulties likely to be encountered or material required and 
the arrangement of all matters so as to facilitate the carrying out of the commander's 

Thus it may be said that, though he commands no one, he assists the commander to 
command everyone, that while he provides nothing but information, yet he arranges 
everything so that the fighting troops and services whose function it is to "execute," 
may be enabled to do so with the maximum of unity and the minimum of friction. 

10. Difficulty in Understanding Staff Principle 

Civil administrators have found it difficult to understand this special relationship 
by which the commander retains his full responsibility and authority while delegating 
much of the work of administration. Co-ordination has always been associated with 
command, with control. It has not been regarded merely as one of the special aspects 
of command, but as bound up with the personality of the chief. Leaders themselves 
have not appreciated the possibility of delegating not only the drafting of formal 
arrangements, but many of the personal contacts and adjustments, necessary to secure 
co-ordination, while at the same time retaining full personal responsibility and 

29 Quotations from British "Field Service Regulations," Vol. I, Organization and Administration. 


Subordinates equally have associated the formal operation of issuing instructions 
with superiority of status. Where private secretaries or assistants are employed by 
chiefs in civil life their relations with subordinates of equal or superior status are 
frequently vague and easily strained. A secretary or assistant who issues instructions 
on his own initiative and signs them is necessarily regarded as "encroaching" on the 
independence or authority of officials senior to himself. The conception that the 
authority and responsibility remain his chief's is absent. Much of this difficulty is 
avoided in military life through the separation of function from status by the device of 
rank. Status is determined by rank and staff officers are almost invariably junior in 
rank to their commander's principal subordinates and the equals in rank of the 
chiefs of services whose work they co-ordinate. Thus it is clear that their actions are 
in virtue of their appointment, their functions, and involve no assertion of unjustified 

Any lack of courtesy, any unreasonableness in instructions, can be taken up 
immediately by the subordinate with the commander. The commander on his side will 
know, or will find out, whether the difficulty is due to the fault of the staff officer or to 
a failure on the part of the subordinate, perhaps unavoidable, to distinguish between a 
local and a general situation. On the other hand the fact that the chief is himself 
responsible for the staff officer's action precludes any wide possibility of pursuing an 
unfortunate personal relationship through such channels. The subordinate knows that 
unjustifiable criticism of the staff will react to his own disadvantage. The staff officer 
knows that failure to make the best arrangements possible in any given circumstances 
will probably be reported to his commander. 

Failure to appreciate the proper functions of officials in a "staff" capacity or to 
recognize the necessity for this third type of control, distinct both from "line" and from 
"specialized" authority, is one of the chief obstacles to more effective co-ordination in 
civil administration. Co-ordination depends in large measure on personal relations, 
reinforcing detailed and definite provisions for securing correlation at every point 
where it is necessary for effective effort. The chiefs of large modern enterprises, cannot 
and should not have time, either to work out the details which follow from their deci- 
sions or to explain them personally. The heads of the major functions and departments 
of such enterprises have their own heavy executive responsibilities. Moreover, their 
attempts at self-co-ordination are inevitably biassed and consequently lead to much 
unnecessary friction. Where it is no one's business to co-ordinate, except a chief's, who 
has a thousand other preoccupations, it is a miracle that some unity of action is never- 
theless achieved. 

1 1 . Staff Relations 

In view of the importance of the subject it is suggested that some closer study of 
military practice in this matter is desirable. 

Figure V shows in diagram the relations between a British staff officer in the field 
and those in immediate touch with him. He is under the "line" command of his chief. 
In virtue of his position he issues instructions to his chief's immediate subordinates 
who are his superiors in rank and to the heads of specialized services under his chief's 
command, who are his equals. He communicates officially through his chief and his 


t.^t « 

• rtfl IBM « 

<a *» « a c a a 

O 3 H*» BBS 

S3 O Vi fl h -4 

S 3 o a i w. j- 



1— I 








I 3 

fcH wo~.e 

o 3 oi a-H 

- -3 o o ~» --' 




The Oeneral with & through 
faia Staff 



I G.3. 



















who with & 
through hie 



















1 1 




who with k 
through Officers 
performing etaff 


special late 


who with & through 
Officers t NCOs. 
acting as staff 





INTELLIGENCE B0 "?J"° S *?!!r« 






chief's superior with that superior's staff officer of the same branch and the heads of 
specialist services under his command. He communicates officially through his chief's 
immediate subordinate with that subordinate's staff officer of the same branch. These 
"channels of communication" safeguard "the scalar process." 

But they are supplemented by a network of personal contacts of all kinds between 
the staff officer and other staff officers of the same branch above and below him, the 
heads of specialist services above and below him, and the commanders of fighting 
troops whom he serves. The minute such personal contacts degenerate into personal- 
ities, anyone concerned can get back into the "official channels." But everyone con- 
cerned also knows that the official channels are slow and the necessities of war urgent. 
The importance of these personal contacts as supplementing and expediting official 
procedure is a feature which is emphasized in all staff work. If there are no officials 
whose specific function it is to secure such liaison, misunderstanding and friction are 
almost inevitable in any large organization. 

Figure VI shows these relations as part of the general structure of command — 
"the scalar process." Each commander and his "staff" operate as a unit, the authority 
and responsibility remaining the commander's. Specialist or "functional" troops and 
services have to be commanded just as much as "line" formations and units. But, while 
nominally they report directly to the commander, actually all official business is 
transacted by his staff. He is thus safeguarded from exceeding his "span of control." 
The safeguard is in a sense a fiction or convention. But it works smoothly in practice 
and is seldom abused. Where the command of formations and units of all kinds is thus 
exercised by commanders and their staffs acting as a single personality, there is in 
effect a specialization and segregation into levels of the actual functions of command. 

12. Staff and the Function of Supply 

Figure VII shows how "staff" officers, or officers of the line acting in a "staff" 
capacity, at every level of command secure the co-ordination of a particular service, 
that of supply. 

It must be noted that supply will be one of 12 to 20 specialized services and func- 
tions for which the commander is responsible, and one of five or six major questions 
allotted to a subordinate staff officer of an administrative branch of the staff. But 
at each level of command there is such a staff officer concerned and solely concerned 
to secure co-ordination, both laterally between the supply specialists and the remain- 
ing troops of the formation whom they serve in respect of that function, and up and 
down the line of supply. 

A second point of interest is the fact that the degree of specialization of those 
concerned with supply varies at different levels in the chain of command. At the two 
higher levels (in Army parlance the Corps and the Division) there are officers of a 
special service concerned professionally solely with transport and supply. Such a 
specialist of appropriate status is placed in a "line" relationship to the commander of 
the formation, subject to co-ordination by his "staff." But he is under the technical 
control of the superior officer of his service in the higher formation. The Commander 
of his formation, for instance, would not dream of interfering with him over questions 
of method or the internal economy of his own functional units. The commander's 






concern would be with the results obtained in the supply of his troops. Similarly the 
specialist officer would look to his own service for promotion and personnel matters. 

At the next level of command (the Brigade) there is no specialist supply officer 
in a "line" relation to the Commander. But a unit of the supply service under the 
"line" command of the supply service officer of the Division will be detailed to work 
with each Brigade — during the Great War a Horse Transport Company of the 
Divisional Train. The work of this unit is co-ordinated by the administrative staff 
officer of the Brigade who is in constant touch with its commander. Should the Brigade 
be "detached" from the divisional command, this supply unit automatically comes 
under the "line" command of the commander of the Brigade. 

This difference in arrangements for control corresponds to a difference in the 
physical arrangements for the delivery of supplies. Supplies were loaded on the wag- 
ons of the horse transport companies of the Divisional Train under divisional control, 
ready packed for distribution to units, and were delivered direct to the transport of 
units. The Brigade commander, while remaining responsible that his troops were 
properly supplied, had a less intimate concern with the actual arrangements than 
the commanders of the higher formations. 

At the next lower level (in the British army the Battalion, artillery Brigade, or 
other units) the question of supply is taken over from the specialist service by the 
unit itself. But there is still a substantial degree of specialization. The officer responsi- 
ble is known as the Quartermaster. He is not a trained professional supply officer. 
He is usually a promoted non-commissioned officer of the unit. His duties combine 
the functions of two specialized services — supply (food and forage) and ordnance 
(equipment, clothing, etc.). But he is wholly specialized in these duties and will, in 
practice, devote the remainder of his career to them. He is not responsible for trans- 
port, which is commanded by a combatant officer of the battalion, specialized in this 
duty for a period of years after a preliminary course of instruction, but looking to 
return to ordinary command. 

The commander of a Battalion usually delegates questions of interior economy 
to his Second-in-Command. The Quartermaster acts in a staff capacity to this officer, 
but also exercises "line" command over headquarters personnel concerned with his 
functions. The Quartermaster is of course expected to be conversant with all arrange- 
ments, procedure, and regulations governing questions of supply and to secure that 
his unit benefits from them. 

At the next level of command (the Company) the supply arrangements are again 
specialized in the hands of a non-commissioned officer — the Company Quarter- 
master Sergeant. He carries out the same duties as the Quartermaster and similarly 
acts in a "staff" capacity to the Company Commander or to his Second-in-Command. 
But, like the Transport Officer, he is specialized in such duties only for a period and is 
in the normal line for promotion to other duties with the remaining non-commissioned 
officers of the Battalion. 

At this point specialization of supply arrangements terminates. 

70 the science of administration 

13. Supply in an Industrial Undertaking 

Figure VIII shows, for purposes of comparison, the Purchasing Department 
organization of a world-famous business enterprise with scattered producing units. 
It is interesting for what it omits. 

It does not show how Purchasing is co-ordinated with the central control at the 
top. The report from which it is drawn 30 indicates that in the case of 22 important 
companies investigated the Chief Purchasing Officer reported 

To the President in 4. 

To the Executive Officers (presumably a committee) in 6. 

To the General Manager in 8. 

To the Treasurer in 2. 

To the Superintendent in 2. 

But Purchasing is an essential function of business. If it does not report to the Principal 
Executive Officer how is he to secure co-ordination when purchasing activities are 
required for one of the functions under the control of another subordinate? When 
Purchasing reports to the General Manager and the Treasurer is dissatisfied with the 
stationery or wants 20,000 forms in a hurry, what happens? Usually the Purchasing 
Officer fights it out for himself. Sometimes he goes to the General Manager and he 
deals with the Treasurer. In any event the question reduces itself to one of argument, 
which means relative weight of status and personality, not control and swift executive 
action. If, on the other hand, Purchasing does report to the principal executive officer 
the latter has too many subordinates. 

Secondly the chart does not show what is the relative authority of the Divisional 
Purchasing Agents and of the Managers of those Divisions. The text states that "Divi- 
sion Purchasing Agents, in addition to centralizing the purchasing functions in their 
own particular divisions, are authorized to place directly with vendors orders for 
materials only needed for their plants or orders for emergency supplies which are 
needed at once." But if the Division Manager disagrees with his purchasing agent 
as to what is an emergency, what happens? What are the limits of the technical and 
financial control of the General Purchasing Agent? Such difficulties are bound to 
arise. Who is there to foresee them, to prevent them arising, to deal with them when 
they occur? If the Principal Executive Officer attempts to do so, he will be over- 
worked in a month. 

The point which the chart illustrates is a simple one. Specialization high up in 
the organization of a large enterprise of the control of functions which interlock in 
execution is bound to create throughout the undertaking innumerable interrelation- 
ships of subordinate officials with different functional points of view over problems 
which call for action. If the "scalar process" is clear cut the "line" officer will usually 
have his way and the advantages of specialization will be lost. If it is not clear cut 
there will be a blurring of responsibility and consequent delay or inefficiency in 

Every one of such interrelationships is a point calling for co-ordination. 

30 "Functions of the Purchasing Agent." Pamphlet issued by the Policyholders Service Bureau of the Metropolitan 
Life Insurance Company. 


& ! 


















San Francisco 
Biv. Purchasing 

O i 

O ft © 
M eg » 

o o 5 

h*4 fa 

to > 8 

a p to 



f «0 

s< a 

o *« .. 

n *» 

(B S3 S3 







1 c3 







S S3 





p •* 

w *» 
o cs a 
ece « 
« o c$ 
o M -4 

at <S 

a 3 g 



5 3? "•* 

















Boston Sir. 







ft » 


& c o 

CS CO +» 


4* «J G 

o .c ® 

(8 (0 M 


•-» .m o 

< f\ 

4» Ik 4» 

< «H W 




72 the science of administration 

14. Staff and General Co-ordination 

Turning from the analysis of a particular function to the general co-ordination 
of a large undertaking, Figure IX shows the co-ordination of all functions in a British 
infantry division as it was organized during the Great War. Some of the details of 
specialist troops and services are omitted. The same principles of organization were 
followed as in the case of the infantry. The points of importance are: 

a. The presence at every level of command of staff officers or of officers or non-commissioned 
officers of the unit concerned acting in a "staff" capacity, i.e. "assisting the com- 
mander in the execution of his functions of command." 

b. The fact that the "scalar process," "line" command, runs clear down from the General 
commanding to the private of infantry. The same is true for privates of other arms and 
services not shown in detail. 

c. That this does not prevent a high degree of specialization or functionalization, more 
than a dozen functions apart from the "line" command of infantry and artillery (why 
not production and selling?) being provided with their own specialist departments or 
officials with considerable though varying degrees of technical independence. 

d. That the method of organizing this specialization varies in the case of every function. 
The degree of professional independence (membership of a separate technical Corps 
with its own regulations and system of promotion) and the level to which this independ- 
ence is carried down in the chain of command is adjusted to the character of the function. 
Each battalion has medical and ordnance personnel belonging to these services. Signal 
personnel of the Royal Corps of Signals are not detached below Brigades. Machine gun 
specialists of the Machine Gun Corps are centralized under Divisional Headquarters. 

e. That, as described in connection with the previous chart, specialization is carried down 
below the levels at which these professional officials are employed by officers and non- 
commissioned officers of the unit concerned who (i) are permanently specialized on a par- 
ticular function or (ii) carry out a specialized function for a period after instruction, but 
expect in due course to be promoted to other duties. 

f. That, despite this multiplication of functions, the line command at every level is enabled 
to secure complete co-ordination of all activities for which it is responsible, without in any 
case exceeding a total of six individuals directly supervised, because, and only because, 
of the presence of the "staff" function. 

g. As a result of the tying together of line and functional specialists by the "staff" func- 
tion, every conceivable service and subject has its clear channel of communication and 
action from the commander of the division to the private soldier. And, at frequent inter- 
vals along those channels, are officials whose express duty is to see that they work and to 
take immediate and direct action if they do not. 

15. General Co-ordination in an Industrial Undertaking 

Again for purposes of comparison, Figure X illustrates the principles of organiza- 
tion of the General Motors Export Company. 31 

The four General Managers at headquarters are spoken of as being in a "staff" 
relationship to the General Manager. In fact they are specialized officers in "line" 
control of particular functions. In the absence of any "staff" system in the military 
sense the amount of specialization is necessarily limited to these four functions; 

31 Reproduced from "Organization and Operating Principles," Edgar W. Smith, Handbook of Business Administra- 
tion, pp. 1474-1488. 


Officers Ic NCOe. of the "Line" 
Unfantry & Artillery • 

Officers & NCOs. of specialist 
troops fc Services O 

Do. t Wh<»»» specialized fV 

Do. .when employed in "Staff" 
duties ^£ 

Staff Officers & NCOs. A 

tine command ........ , 

DOllW iti "s.taff" coordination 

Technical Direction.. 

Staff instruction ... — . 

Staff coordination •• 

Staff Liaison -„ 











8 8 2 

I I 

! i 

i i 






• ■ 





































\ \ 


















f I 



































































— 1 






otherwise the General Manager's "span of control" would be exceeded. The number 
of Regional Directors is not shown. 

It should be added that this represents a very advanced form of business organiza- 
tion which has been developed with exceptional insight. But it is doubtful if the two- 
dimensional grouping of authority and responsibility on which it is avowedly based 
is sufficiently refined for the requirements of modern scientific administration. The 
"Line and Staff" form of organization might be renamed the "Line and Functional" 
form. It would then appear for the compromise which, essentially, it is. It would be 
replaced in due course by a "Line, Staff and Functional" form, "staff" in the mili- 
tary sense being introduced to effect a real integration between the other two prin- 

16. The True "Staff" Concept in Business 

The only positions in civil life which at all correspond to those of "staff officers" 
in the Army, are Private Secretary and Executive Assistant. While there is great 
variety in practice and some "staff" functions inevitably gravitate to such officials, 
it is clear that they are seldom, if ever, used solely or in large measure to secure 
devolution of the actual functions of command. The degree to which "staff" are 
allowed to assist their chief is largely an individual question, depending on the chief's 
powers of delegation. 

One enquiry directed to companies including the position of Executive Assistant 
in their organization revealed an astonishing variety of practice. 32 The Assistants in 
one case or another exercised direct supervision over nine different types of function 
or department. They acted as Secretaries or members of nine separate kinds of com- 
mittee. They performed a heterogeneous collection of special duties from "a survey 
of the general management methods of the Company" to "assistance in connection 
with some special hobby or interest of the superior." Finally they performed co- 
ordinating duties or served in a liaison capacity between "their chief and various 
officials" — the wording implying that, where this occurred, it was a special function 
requiring definition in each case. There was no indication at all of any general recog- 
nition that the central purpose of officials in a true staff capacity is to assist in co- 

The report concluded: 

"the position of an Executive Assistant has come about primarily because of the multiplicity 
and urgency of the demands on the time of the chief executive in present-day business. It is 
the principal function of the assistant to help to relieve this pressure." 

These words may be compared with those of the author who may be described as the 
father of the modern general staff theory in military organization: 

"As long as armies were small, and movements, encampments and fighting formations were 
laid down by hard and fast regulations, the want of trained General Staff officers was less 
felt. The plan determined upon by the General in command usually contained the details of 
execution. . . . The enormous numerical strength of modern armies and the way they 
must be organized to meet the constantly changing requirements of war, render necessary 

32 "Functions of an Assistant to the President." Pamphlet issued by the Policyholders Service Bureau of the Metro- 
politan Life Insurance Company. 


great differences in carrying out the details of military operations even under apparently 
similar circumstances of time and place. Consequently the higher leaders and Command- 
ers require a regular staff of specially selected and trained officers." 3S 

Today, the central point of the pressure on the chief, in business as in military life, is 
the additional work of co-ordination called for by the specialization rendered essential 
by scientific development. The assistant can only relieve pressure from this cause if 
his duties and his relationships with other officials are so laid down as to secure dele- 
gation of actual co-ordinating work. 

17. The Staff Principle and the Span of Control 

There is no question that effective control of large and complex formations such 
as are found in armies would be impossible without "staff" assistance. But it may be 
asked how does the "staff" principle reduce the commander's task of co-ordination: 
does it not merely add further to the number of individuals he has to consider? 

It reduces it in two ways. In the first place it throws upon the staff all the detail 
of commanding. If orders are to be issued, instructions prepared, arrangements 
made for this or that or the other purpose the Commander has only to indicate the 
broad outlines of his decision — with or without preliminary conference with his 
principal subordinates, conference at which his principal staff officer or two principal 
staff officers are always present. At that point it becomes the duty of the staff to work 
out all the consequences in consultation with everyone concerned, to prepare orders 
and instructions, and to satisfy themselves that such orders and instructions can 
be executed. 

In the second place it throws upon the staff much of the detail of controlling — 
of satisfying the Commander that instructions issued are in fact being carried out. 
This part of the work of a staff includes functions both of inspection and of co- 
ordination. The staff officer must satisfy himself as to the facts. If instructions are not 
being executed he must take steps to iron out the obstacles, whether personal or 
material, which are preventing the fulfilment of the Commander's intentions. 

On the other hand it in no way interferes with the "scalar process," or over- 
strains the Commander's "span of control." The commander of a British division in 
the field actually supervises directly the commanders of his three Infantry Brigades, 
and the commander of his Artillery, — his principal subordinates, — and two chief 
staff officers, one for operations and one for administration, six persons in all. Six may 
seem a large number, but the work of the three infantry Brigades does not interrelate 
closely under normal conditions and they are standardized formations. 

He has also under his direct command, it is true, the chief representatives of 
three or four other arms and of six or eight services. Each of these men can be called 
upon for specialized advice if required: each of them has the right to approach the 
commander directly upon an issue of major importance. But 99 per cent of their work 
is correlated by the staff. Just because this is so, and because the relationship is well 
understood, the commander is free to visit units and formations, to see for himself 

33 "The Duties of the General Staff," General Bronsart von Schellendorf. English translation of the Fourth Edition, 
London 1905, p. 4; also quoted by the late Lord Haldane of Cloan in a Memorandum announcing the formation of a 
General Staff for the British Army, London Times, September 13, 1906. 


what is happening, to meet men unofficially and to weigh them up, to stimulate or to 
restrain them, in short to exercise the positive and vital functions of leadership. 

18. The Functions of Administration and the Use of Staff 

It may be suggested that, while the special staff functions and relationships 
which have been described are applicable to armies where discipline is enforceable, 
they are not utilizable in business or other forms of civil administration. No doubt a 
considerable educational effort is necessary before any new concept of organization 
can be introduced into an environment where individuals are unaccustomed to it. 
But, there is no underlying reason to justify the assumption that the principle is not 
equally applicable, and likely to prove beneficial to, other forms of undertaking, 
once it is accepted and understood. 

To take an example from an industrialist of wide experience, Fayol has analyzed 
the functions of administration, of the commander, under five heads: to plan, or- 
ganize, command, co-ordinate and control. 34 But these functions, since they are an 
integral part of the duty of the leader, cannot in any case be separated wholly from 
his personality. A planning room can secure that the routine operations of manufac- 
turing are systematically performed and that all minor obstructions or mishaps are 
dealt with. It cannot settle the policy of a business, forecast its probable lines of future 
development, or make major decisions. The responsibility for preplanning in this 
larger sense (Fayol' s word was "prevoyance") 36 must remain with the highest 

Organization must depend on such preplanning. It is the provision and arrange- 
ment of the appropriate human and material resources for a given purpose. That pur- 
pose depends on the plan. Only he who is responsible for the conception and broad 
outlines of the plan can really decide in the last instance whether such resources are 
appropriate, excessive, or insufficient for the purpose. To command is to set the or- 
ganization to work in accordance with the plan. Every subordinate who exercises 
delegated authority commands. But that authority is delegated from somewhere. 
Finally the function of command must lead back to the principal executive officer; 
the responsibility cannot be divorced from his personality. 

To co-ordinate is described as "to give things and actions their proper proportions 
and to adapt the means to the end." 36 In other words, it is the operating side of 
organization. Where organization is concerned with quantities and numbers and 
the setting up of a structure to facilitate their unified working, co-ordination is 
concerned with securing that working from day to day and hour to hour; it involves 
constant attention to the machine in action to secure that harmony and balance are 
preserved in operation as well as in design and structure. 

Control consists in seeing that everything is carried out in accordance with the 
plan which has been adopted, the organization which has been set up, and the 
orders which have been given. Maintaining the metaphor of the machine, it is the 

34 Henri Fayol, "Industrial and General Administration." English translation published by the International Man- 
agement Institute, Geneva. 

35 "To foresee," as the word is used here, means both to foretell the future and to prepare for it; it includes, in fact, 
the idea of action, and can best be expressed by "planning." Ibid., p. 35. 36 Ibid. 



gauges and records of performance. Just as co-ordination is the operating side of 
organization and depends on it, so control is in a sense the consequent of command, 
the constant checking up of the results of command in action. The commander cannot 
delegate wholly his personal responsibility for co-ordination and control, for the same 
reasons that he cannot delegate it in matters of organization and command. 

If Fayol's analysis is examined closely it appears that his idea of planning also 
contains two related conceptions, one involving the "foretelling of the future" and 
the other the consequent action of "preparing for it." If, at the same time, investiga- 
tion is given its proper place as the underlying principle in all decisions in any scien- 
tific form of administration, Fayol's arrangement of the functions of administration 
corresponds closely with Mooney and Reiley's "Principle, Process and Effect" in 
their arrangement of the principles of organization. Moreover, these functions fall 
into three groups corresponding to the three aspects of staff service which Mooney 
and Reiley distinguish, viz., the Informational, the Advisory, and the Supervisory. 




The Functions of Command 




(Fayol's analysis in italics) 










Aspects of staff service 




(Mooney and Reiley) 

The principle of investigation leads to the process of "foretelling the future" 
and the effect of "preparing for it." But "foretelling the future" issues into process 
with the establishment of organization and organization in use has the effect of co- 
ordination. Similarly the plan enters into process through the action of command 
which has the effect of control. In investigation the commander needs information. 
In attempting to forecast or to set up his organization and secure its unified working 
he needs counsel. In securing that his plan is worked out in all its details, that his 
orders and instructions reach all concerned, and that they are carried out, he needs 
help in supervision. 

Investigation, as in the case of the other functions of the administrator, cannot 
be wholly separated from his personality. Despite the insistence of the literature of 
business administration on the investigational aspect of "staff" functions, somewhere 
and somewhen the final authority, the man responsible, must take the available facts 
into account. However predigested they may be, it is his brain which must review 
them, weigh them, and base decisions upon them. 

Thus, of the whole seven functions of administration, none can be entirely dele- 
gated to others. The hard core of responsibility remains. At the same time the "staff" 
system, as it has been described, does follow very closely Taylor's main principle 
that one man should as far as possible confine himself to one function. 

As far as the commander is concerned, it does this by enabling him to concen- 
trate as far as is humanly possible, upon the principal of his seven functions, namely 
to command. Fayol lists the main requirements of command in industry as follows: 











2 3 




Q - 
< s 

O -g 
co ■§• 

z "1 

9 I 








"Any man who has to command must: 

1 . Have a thorough knowledge of his staff. 

2. Eliminate the incompetent. 

3. Have a sound knowledge of the agreements between the undertaking and its em- 

4. Set a good example. 

5. Make periodical examinations of the organization, with the help of charts. 

6. Collect his principal assistants in conferences, in which unity of management and 
effort can be arranged. 

7. Not let himself be absorbed by details. 

8. See that his staff possesses energy, initiative and loyalty." 37 

Of these eight requirements, 1, 2, 4 and 6 are essentially matters of face-to-face 
leadership. They cannot, with the possible exception of 6, be met in an office. And 
conferences summoned too frequently at the chief's office, take men away from 
the scene of action: they are less fruitful than conferences on the spot. Number 8 
depends also on personal contacts. If the word staff is used in the wider industrial 
sense this is obvious. If it is used in the narrower army sense, there is no way of 
judging the work of a "staff' so accurately, as frequent personal inspection of the 
results in the comfort and convenience of the administered — the "line." 

In army life 3 and 5 are covered by regulations and War Establishments. 
The leader is little concerned with them unless there are important changes. It is 
the staff's business to know the detail and to advise when required to do so. 7 is 
obviously a condition on which the successful fulfilment of 1 to 6 and 8 depends, but 
it is also applicable to the method of fulfilling the remaining requirements. In any 
event it is likely to become serious, if and when the chief has no adequate machinery 
for delegating the other functions of administration. 

The point of importance is that the essence of the function of command is per- 
sonal contact with all concerned. It is not an office job. The constant complaints in 
industry as to the loss of "the personal touch" mean one of two things. Either the 
administrator is glued to his office doing what should be "staff" duties, or he has no 
personality. Under the Army System of organization no general who uses his "staff" 
properly and has some personality has any difficulty in making it felt throughout 
commands larger than any at present contemplated by industry. 

Since there is no practical experience of the application of the true "staff" 
concept to business undertakings, it is not possible to give any precise measurement 
of the degree to which its effective employment can relieve the chief. If, however, 
Fayol's analysis of the task of administration is accepted, it is possible to indicate 
how far each of the tasks under each of the administrative functions, would, on the 
basis of military experience, normally be carried by Staff Officers. The results of 
such a comparison are shown in Figure XL The proportions of staff service indicated 
represent merely the author's judgment. 

19. The Staff Concept and Function alization 

It is also possible to test the validity of the "staff" concept for business under- 
takings, by comparing military practice in the matter with the principles suggested 

37 Ibid., 70. "Staff" in this context means all employees. 


(Developed from 'AftrintottaWflB laftajWalll at general*' Vr Henri Payed, E n gl l ah tranalation by J.A.Coubrough under 
the title 'Ladu*ixl*l *nd General iVn'.T.Hftn' puhliabed by The International Ui.-_^-ne=t Institute, Geneva: 
obtalnahl* from 31r I*a*o Pitenn and Son Ltd.,L<cdce 

jr.TZZ Fayol'e five function* «f adnlnlstrBtlon have been expanded to arm by dividing ' PrvTOjana* ' into Fcreoaating 

1 , jnd Pluming and e^<»c the principle of Investigation which But underlie an/ solan tlflo malyaia of 
aomlnlatr a tlon. 

2. With thee* addition* Payol'* function* fall Into a logical aeheme of princlpl*,prooeB* and effect ■JTJl tf to t, 
the arrangaoent of the principlea of crganlaatlen developed fey Hocney and Bailey. 

The chart shew* In inverted ocean* the detailed requirement* of each function a* stated by Feyol and, .gainst 
»•* Mouireannt.the way in which ataff can relieve the ohief.Th* ahaded portion of aaoh parallelogram 
Indicates tha proportion of aaoh reauirocnt whioh,lr. tha opinion of the authox.can be uiuad by a nil- 
organised and oompetant etaff. 

Staff ©annot relieve tha chief of final reaponsibility.In all funotiona and requlraaenta therefore 10* of the 
wort has bean ebon aa the ohiersteven la oaaea where the whole routine should be undertaken by the *t*ff, 
tha ohief Barely aaa n mi n g reeponalMllty for aotirltlaa which are new In fact oalled to hie attention, unlaw 
ther are neglected or Incorrectly performed. 





All work Of 

■hould be lone by tha 
ataff 1*. the first 
instance, the ohief 
oonf ining himself 
to collating their 
conclusion* and 
oonf irmlng them by 
personal enquiry and 
/er inspection, wnly 
on laaueo of major 


"la baaed on 

*.The roBouroes of the undertaking 

b«The nature 4 importanoe of the 
operations In hand 

o.Puturo poaoibilltlee which dep- 

-end cm oondltiona all of 

then subject to chengee,who*a 
ljmiartanoe and time of occurr- 
-onee cannot be determined In 
advance •' 

A ataff *hould:- 

a.Belinv* the ohief of all Intellectual preoccupation oy 
assuring oomplete and accurate detaila of all reeouroea 
presented in ocnvsnient form at any tin*. 

b. Present all infarmation and materiel eseentlal to e 
Judgement on thia point. It remains ho wu i u r a Batter af 
decision fr om which they cannot entirely relieve the ohief 
. .■ -!■■ ■■■■ ■■■.:■ to calculate future oonditiona a* far as 
possible and prepare plana In advance, at all event* in 
outline, designed to meet all possible eventualities* 
The chief should have only to select from *mcng varioua 
alternatives, the preliminary wor k of translating various 
combination* of oonditiona Into detailed plans of action 
having been dene for h<m. 

"The general characteristics I 
a good plan of operations an 
a. Unity 

It ia the duty of a etaff;- 

a.To ensure that instructions Issued 
to those under the command of their 
chief are mutually consistent . 
b.To secure that all standing plans 
■-.-.. a Defence Scheme) are kept 
up to date 
c.To csll tbo chief's attention to 
possible contingencies not provided 
for by the plan:but,ainoo thle 
quality Is inherent In the plan it- 
self, It depends principally on the 
Judgment of tha ohief. 
d.To frame plans accurately A eonolaely, 
eliminating detailed lnatruotlans on 
points whsre foresight cannot penetrate. 
Sinoe the staff draft all plena thia 
quality, aubject to any tendency to unoert 
-ainty or ovorelaboratlon on the part 
of the chief, is their responsibility. 
In British Army regulations It lo laid down that 'the will of thw 
Comaander la expressed by his plan' which 'oust be conceived In socordance 
with Ihe established principles of war* i the first of these principles 
is Concentration. 

d •Precision.'' 


"In every cone the organ! ration has 
to fulfil the following admlnis- 
-tretive duties: 
e*Se* that the plan of operation* 

la carefully prepared and 

strictly carried out. 
b.See that the human and material 

organization ore suitable for 

the objects, re ■ouroeo and need' 

of the organization. 

e*ZBtabllsh a management which is 
exxnpetent and vigorous and has 
singleness of purpose-. 

o-i .' !■ decisions which are dear 
distinct and precise. 

f Jluke careful selection of stoff._ 
gtDeflno, duties clearly. 

h, Encourage the desire for 

Initiative and responsibility. 

a ,Cartj in preparation la alacst wholly a ataff responsibility. 
Supervision of execution la partly o ataff duty. Staff 
officere should nets any deviation from plan and ore 
responsible for establishing ayatsms of ooraminiont ion and 
report to keep their chief In tpuch with a changing 

The organiEation of the training of personnel is one of the 
branches of Btaff duties. In any given operation it Is one 
of the first tank* of the administrative staff to see that 
the neoeoaory material ia availahle.On the other hand 
of Force is tha second principle of 1* 
equally the duty of the staff to provide only tha oilnlrani 
of nan and material neoessary for any purpose. 

a. The first two requirements are a matter of selecting sub- 

I- ordinate*, a responsibility which the chief oannot delegate- 
The third is, in part, a question of coordination - a etaff 
re Bponaibill ty, but wor% of 'morale'.'Uorale' depends on tha 
personality of the chief.The staff only affect it indireot- 
-ly their own attitude, ii. by freeing the ohief so that 
he can maintain personal oontact with Bubordinataa. 
d.Wholly a ataff function. 

e.Xhia depends primarily on the chief, but the staff in 
| framing his orders and instruction* exercise a great 

| influence on clarity. 
f*A function of the chief. 

Ic. Reward men fairly end Judic- 
-loualy for their services. 

ltlmpoae pecslties for mistake* 
and blunders. 

n.Swe that individual interests do 
not interfere with the interests' 
of the concern as a whole. 

o.Poy special attention to unity 
of Qoranand 

p,Dnaure material and buaan order 

q,.SubJect everything to control. 
r«Avold red tape." 

g.A staff function. 

i h. Primarily a matter of the chief* attitude, but the way in 

which staff duties ore performed must refloot it. 

k.In a large organisation this can only be secured by routine 
arrange^ento for which the responsibility must be very 
largely etaff. 

l.The routine of discipline is a staff responsibility. 

•A responsibility which is primarily the chiaf'Sjbut in 
which he is usually assisted by a special staff officer. 

n.It is the function of the ataff to saa that •dminiatoBtlao 

is 0T-dtahle 

Part of the ataff function of coordination* 

.Largely a etaff function associated with coordination and 

q.Effective control depend* on timely presentation of approp- 
-rlate and accurate Information and *tatl*xles:their design, 
collation and presentation are a ataff function. 

r.It ie a duty of all staff* to reduce corresponnence and 

procedure to i 

Since it Is with the funotion of Command that the chief's plan enters into prooee* the variou* eapeot* of thia; funot- 
-ion are those which depend moat intimately upon the pergonal qualities of the administrator .They oazsart ba delegated 
to ataff In the same proportion as the other eapeot* and funotiona of administration* The main duty of *tsff In relation 
to them 1b to free the chief from preoccupation with the other aspect* and function* so that he may concentrate am 

"Uiy man who has to < 
a*Bave a thorough knowledge 

his staff, 
b, Eliminate the incompetent. 

0*Hsre a Binond knowledge of the 
agreements between the under- 
-taking and its eimployees. 

cUSct a good example. 

etliake periijdioal examinations 
of the organisation with the 
help of charts. 

f.Collect his principal asaie- 
-ta&rt* in conference b, in which 
unity of management and effort 
can be arranged, 

g.tfot let himself be absorbed by 

h.S*e that his staff possesses 
energy, Initiative and loyalty 

a. Is primarily a matter of opportunities for personal aontaet 
and Judgement. Insof ar as It may depend on advice or ok 
records it is a staff function. 

b.Thia Is entirely a matter for the chief's Judgement. 3 toff 
assistance Is confined to securing that formal procedure 
designed to secure iumartiallty,to protect the individual 
or to safeguard the chief from subsequent diapute or 
criticism, la strictly observed. 

clt is the function of the staff to have knowledge and to 
maintain detailed records of all regulations, a g ree m ents 
and understanding* bearing an the terms of service of 
and relations with employee*. They should advise their 
chief on the bearing of such data on each individual case. 
The chief should have such general Icnowledge of principles 
and practices ■ --<11 enable him to question any 
to hasty or Incuoslstent Judgeaents an the part of the 

d.7his ia wholly a matter for the cfaiafjfl* *taff,aa rep- 
resenting hiOfhave (=..-. a special personal obligation 
to assist him in this respect. 

a. Staff should maintain appropriate record* for thia purp- 
-cse.coll the chief's attention immediately to any 
1. departure from the principles on which the organJ ration 
ia fr ^m^* ^ t << ,wH mmH» i-n-T n p^<ng a* to duties and respozt- 
-sibilitie* or iii.exces* of peraonnel over budget* cr 
-establishment* and prepare plans and figure* for hi* 
approval whenever modification of the org anl cation aeemsi 

f.Staff should arrange such conferences so that they 

Involve ti» ndnJaajB interference with normal work, assist 
at then and conramicate their results to all concerned. 

g»Depends on the chief's peraonal capacity to delegate, but 
efficiency of staff work determineB the degree to which 
delegation is possible. 

atter of peraonal supervision by the chief. 


"To coordinate 1b to harmonise "11 the operations of a concern eo as to leal 
to the anooth working which makea for Is to give the right 
proportlcms to the material end tamian organisation of each function, «e that 
it may play its part with certainty and id low for the need* and 
consequences which every operation .... will Impose upon all the funotiona 
of the oonoern;to proportion .... erpendlture to reeouroea, alee .... to 

nee da, re serve a .... to oonsumptian precautions to dangorajto make 

the important point take precedence over la In a eBntenoe te 
give things and aotiens their proper proportloru) and to adapt the man* to 
tike end. 

In any organization which 1b well coordinated we find the following oondlt- 
-lons:a.Eaah department work* in harmony with the others: .... every 

operation i* osrrled out with order and certainty, 
b. Within each departanent the varioua sections and subaeotlon* are 

given exact instruction* e* to the part which Ubj have to play in 

the ooramon work and how they mist oociblne with each other. 
o.The progromse of work for each deportment and section 1* always 

kept up to date." 

Coordination thus Involves alone planning of work to lnolule *U functions^ 
including all neoeanary calculation* in term* of time, place and quantity, 
erramgil nation of Instruction* to all concerned, *uperrl*ion to secure that 
they are understood and their constant adaptation to circumstance a. 
It Is the at aff function par ex nellenoe . 


"The control of an undertaking oansista In neelng that everything la being 
carried out in accordance with the plan whloh has bean adopted, the order* 
which have bean given and tha prlnoiplea whloh have been laid* 
object la to point out Bd*t*k** In order that they may be rectified and 
prevented from recurring again.Control oust be applied to everything whloh 
farms part of tha undertakln*^ to men,materials and operation*.' 1 

It is an oBser-tlal function of a etaff to provide their chief with all the 
information necessary to the eXBrciee of continuous oantrol by designing, 
o oil at lng, checking and analysing an appropriate aerie* of routine report*, 
returns and atatistlcs,and the insure both their accuracy and timely 
Information a* to immediate situation* by constant personal Inspection, 

(Developed from the work of Henri Fayol) 


"Any man who has to command must: 

1 . Have a thorough knowledge of his staff. 

2. Eliminate the incompetent. 

3. Have a sound knowledge of the agreements between the undertaking and its em- 

4. Set a good example. 

5. Make periodical examinations of the organization, with the help of charts. 

6. Collect his principal assistants in conferences, in which unity of management and 
effort can be arranged. 

7. Not let himself be absorbed by details. 

8. See that his staff possesses energy, initiative and loyalty." 37 

Of these eight requirements, 1, 2, 4 and 6 are essentially matters of face-to-face 
leadership. They cannot, with the possible exception of 6, be met in an office. And 
conferences summoned too frequently at the chief's office, take men away from 
the scene of action: they are less fruitful than conferences on the spot. Number 8 
depends also on personal contacts. If the word staff is used in the wider industrial 
sense this is obvious. If it is used in the narrower army sense, there is no way of 
judging the work of a "staff" so accurately, as frequent personal inspection of the 
results in the comfort and convenience of the administered — the "line." 

In army life 3 and 5 are covered by regulations and War Establishments. 
The leader is little concerned with them unless there are important changes. It is 
the staff's business to know the detail and to advise when required to do so. 7 is 
obviously a condition on which the successful fulfilment of 1 to 6 and 8 depends, but 
it is also applicable to the method of fulfilling the remaining requirements. In any 
event it is likely to become serious, if and when the chief has no adequate machinery 
for delegating the other functions of administration. 

The point of importance is that the essence of the function of command is per- 
sonal contact with all concerned. It is not an office job. The constant complaints in 
industry as to the loss of "the personal touch" mean one of two things. Either the 
administrator is glued to his office doing what should be "staff" duties, or he has no 
personality. Under the Army System of organization no general who uses his "staff" 
properly and has some personality has any difficulty in making it felt throughout 
commands larger than any at present contemplated by industry. 

Since there is no practical experience of the application of the true "staff" 
concept to business undertakings, it is not possible to give any precise measurement 
of the degree to which its effective employment can relieve the chief. If, however, 
Fayol's analysis of the task of administration is accepted, it is possible to indicate 
how far each of the tasks under each of the administrative functions, would, on the 
basis of military experience, normally be carried by Staff Officers. The results of 
such a comparison are shown in Figure XL The proportions of staff service indicated 
represent merely the author's judgment. 

19. The Staff Concept and Function alization 

It is also possible to test the validity of the "staff" concept for business under- 
takings, by comparing military practice in the matter with the principles suggested 

37 Ibid., 70. "Staff" in this context means all employees. 


by F. W. Taylor. In British army organization complete functionalization in Taylor's 
sense is not carried below the command of a Company (250 men). Below that level 
his description of the old form of industrial organization as "military" is partially 
applicable. Even at the next lower level, however, the Platoon (20-40 men), the 
Commander has a Platoon Sergeant, who acts in a staff capacity, relieving him of the 
detail of the interior economy of the platoon, and enabling him to concentrate on 
its leadership, training, and fighting efficiency — a greater degree of functionalization 
than is found in industrial enterprises not touched by Taylor's ideas. The comparative 
uniformity of the private soldier's duties and the moral necessity of associating him 
by every possible tie to the man who will actually lead him in battle, make function- 
alization at the lowest levels of command unnecessary and undesirable. Moreover in 
many matters he does receive the direct attention of functional specialists at his 
Company and Battalion headquarters. His Platoon Officer and Sergeant have a far 
more limited overall control than the old-fashioned foreman in industry. 

There is, however, no reason why, under different conditions, functional control 
should not be carried right down to the individual if it is desirable to do so. Organiza- 
tion in the British Army could be so adapted merely by changing the duties of ex- 
isting personnel. The Company Commander would become the "Shop Superintend- 
ent" and his Second-in-Command the "Assistant Shop Superintendent." The duties 
of the four Planning Room specialists might be allocated to the four Platoon Ser- 
geants, and the four Platoon Officers become the foremen in the shop. They might 
have two groups of eight assistant functional foremen drawn from the Non-Commis- 
sioned Officers in charge of Sections. 

But this arrangement would not offer any solution of the problem raised by 
functionalization carried to the higher levels of organization. Somewhere or other 
there must be co-ordinating points, especially when subordinate units are geograph- 
ically separated. It is the extent of authority given to the manager at such points 
and its relation to functional authority which are in question. 

As far as the "staff" itself is concerned, it is divided into branches which imply 
functionalization, at all events for the time being. While Staff Officers are not special- 
ists, they carry out special duties. In the British Army there is first the broad division 
into the General Staff Branch and the three branches of the Administrative staff 
— dealing with: 

Personnel (Adjutant-General's Branch). 

Supplies, animals and certain classes of stores (Quartermaster-General's Branch) and with 
Equipment, both maintenance and research (Branch of the Master General of the Ord- 

respectively. These branches are again subdivided into sections. For example the 
General Staff Branch has sections for intelligence, operations, and training. This 
specialization within the staff is carried as far as possible, compatible with the staff's 
underlying function of co-ordination and the employment of the minimum number of 
persons necessary for the work. 

In this way certain of the main aspects of administration are very highly func- 
tionalized in certain branches or sections. Investigation, which is most important in 
war in relation to enemy action (cf. the competitors' activities aspect of market re- 


search), is dealt with by the Intelligence. Forecasting is expressly the duty of the 
General Staff Branch. Organization on its formal side is a matter for the Adjutant 
General. Planning, in its broader aspects, is essentially a question for the Operations 
Section of the General Staff, though all branches and the specialized troops and serv- 
ices contribute their detailed aspects of a complete plan. Control on its systematic 
and formal side is largely confined to the Adjutant General's Branch. The accounts 
of war are balances in terms of men. Returns of strengths, casualties and prisoners are 
the basis of such accounting. But each section and branch assists in control, by sta- 
tistics of the matters with which it is concerned and by constant personal visits to sub- 
ordinate formations and units. 

Command is the function of administration which it is most difficult for the chief 
to delegate. But the operative instruments of command are orders and instructions. 
All the formal work of preparing and issuing such orders and instructions is assumed 
by the staff, the work being usually functionalized into the two main groups of 
General Staff and Administration. Co-ordination, the fundamental reason for staff 
organization cannot in its nature be functionalized. It is provided for by the scalar 
structure of the staff itself, and in the British Army by a special arrangement under 
which the commander while 

"responsible for co-ordinating the work of his staff . . . may, on occasions, delegate it, 
with such reservations as he may think fit, to his senior general staff officer." 38 

Thus the duties of the staff are divided up and grouped by functions in a very 
large measure. In certain matters further functionalization is possible and probably 
desirable, e.g. in the control of clerical work. On the whole, however, Taylor's prin- 
ciple of one man, one function, is followed far more closely than in the majority of 
civil enterprises. His description of the older form of industrial organization as 
"military" does not correspond with the facts. 

But it is the functionalization of levels of authority, the actual division of the 
work of administration, so that the chief is relieved of everything except its most 
personal aspects, the primary initiatives and the vital decisions on questions of major 
importance, which constitute the chief interest of the staff concept from this stand- 
point. Administration itself implies a number of functions, which can be specialized 
and delegated, provided the authority and responsibility vis-a-vis "line" superiors 
and subordinates, remain with the chief. A plan has many stages. So has its execution. 
The vital decision which sets the plan in motion may be brief in the time that it 
occupies and the form of its expression. But the detailed consequences are both ex- 
acting and intricate. Provided, however, that principal line subordinates understand 
that the vital decision is the chief's and that the detailed arrangements which follow 
are contributory to the plan, there is no reason why they should resist such arrange- 
ments when issued by their chief's "staff" even though those signing the actual 
documents are their inferiors in status. 

38 "Field Service Regulations," Vol. I. 

82 the science of administration 

20. Planning and Performance 

Another of Taylor's main principles where the military parallel is of interest 
is the separation of Planning from Performance. Mooney and Reiley have observed 

"Another common misconception concerning staff service must be disposed of, namely 
that it is in some way specially identified with the planning function. ... If it were really 
the staff that plans and the line that executes, this would elevate the staff above the line in 
functional importance, a reversal of their true relationship. . . . On the Taylor system of 
functional foremanship, . . . every one of the planning bosses and likewise each perform- 
ing boss has the authority of line decision with respect to his own functional job. It is ever 
thus in organization. The line plans, the line executes, the line does everything. In the line 
alone rests the authority to determine plans, the authority to execute such plans, and the 
responsibility for what is done." 39 

This comment appears to miss the essential point which is characteristic of 
"staff" functions in military life. It is true that authority rests in the line alone. It is 
true, as the authors point out, that 

"thinking and doing are inseparable in every psychic movement and are therefore present 
in every activity." 

But in a very real sense it is the "staff" that plans and the "line" that executes. Officers 
in command of line, specialist and service troops plan in respect of their own units 
and functions. As in the case of Taylor's foremen they have authority for line deci- 
sions with respect to their own functional jobs. The special characteristics of the 
"staff" officer's position is that his authority to plan with is that of his line superior, 
not his own, and he has no subordinates other than one or two junior "staff" officers 
and some clerks, to whom to communicate "line" decisions with respect to his own 

A major aspect of the directive function of the chief is planning and the co- 
ordination of specialized plans. The great proportion of this work is delegated to the 
"staff." But the responsibility for execution remains with the chief and the chief's 
line subordinates. By very reason of his special position the "staff" officer, as an 
individual, is the purest example of the segregation of planning from execution which 
can be found. It is a purer example than Taylor's functional foremen, because they 
had authority for "line" decisions in respect of their own functions, as in the case of 
specialist troops and services. 

The difficulties which have been encountered in practice may be traced to a 
single fact. Immediately the scalar claim is lengthened to five or six levels of authority 
it becomes impossible to preserve "line" responsibility and at the same time to co- 
ordinate effectively interrelated functional authorities whose "line" authority for 
their function runs through all these stages. "Staff" relationships are the solution of 
this problem, because they functionalize the planning aspect of command and thus 
present a clearer separation of planning and execution. A good staff somewhat re- 
sembles a Planning Room. With each advance in the technical aspect of war it re- 
sembles it more closely. A modern "creeping barrage" table with defined targets for 

39 Op. cit., pp. 61 and 62. 


every battery and every gun, "lifts" every half-minute, and so on is of the same order 
of detailed prearrangement as an instruction card. But it may be made out by a 
second-lieutenant assisting the "staff" at headquarters as orders to half-a-dozen 
Brigadier Generals. 

Planning in this sense is necessarily anterior to execution in point of time. The 
"line" and specialists cannot act coherently without it. Since the "staff" "adheres" to 
the line, it "adheres" also to the authority of the line commander. In this sense 
"staff" functions precede the functions of officers of the "line" who are under the 
authority of the "line" superior. But this does not mean that "staff" are of greater 
importance than the "line." Their responsibility is greater, because it is more ex- 
tensive. Marshal Foch said to students at the French Staff College before the war: 
"Remember, gentlemen, battles are never lost by the rank and file, always by gen- 
erals." The staff, as part of a general, can help him to lose battles. But in war the 
only two functions of ultimate moment are the executive action of the soldier behind 
a gun or bayonet, and the leadership which puts him in the right place at the right 
time in the right heart. To link these two together, to maintain the one and to protect 
the other from all extraneous and secondary preoccupations, is the end of all inter- 
mediate functions whether of line, staff or service. 

Other principles enunciated by F. W. Taylor also find expression in true "staff" 
organization. It is clear, for instance, from what has been said already that efficient 
"staff" work demands a comprehension and application of the "exception princi- 
ple." It is the custom of the "staff" in the field to present to the commander daily 
statistics of strengths, casualties and other matters with explanations of all unusual 

Taylor insisted that 

"one should be sure, beyond the smallest doubt that what is demanded of the men is entirely 
just and can certainly be accomplished." 40 

British army regulations read: 

"The staff must be in a position to appreciate what is possible." 41 

Taylor stated that in his experience "almost all shops are under-officered." It 
is precisely the lack of officers specialized in "staff" functions which is creating current 
difficulties in co-ordination. 

21. Consequences of Lack of Co-ordination 

Two consequences of the lack of co-ordinating mechanisms in civil organization 
may be noted. First is the proliferation of Committees which is so characteristic of 
undertakings of all kinds. A Committee is a device for correlating different points 
of view. But it is an extremely expensive device. It is sometimes forgotten that the 
cost of any committee is the combined salaries of its members for the time spent at the 
committee or preparing for it. Possibly the popular saying that "the best kind of 
committee is a committee of one" is a subconscious recognition of the fact that many 
committees have only come into existence because an individual vested with the 
authority and possessing the capacity to do co-ordinating work has not been appointed. 

40 "Shop Management." 41 "Field Service Regulations," Vol. I. 


In organizations there is a tendency to compensation, comparable to the tend- 
ency found in the human organism. Where essential mechanisms are not provided or 
decay, somehow and somewhere growth occurs to meet the functions of those mech- 
anisms, often with great waste and at the risk of destroying the whole balance of the 
organism. As Mooney and Reiley have shown, committees may fulfil a genuine 
"staff" function. They may be designed to secure the "infiltration of a true service 
of knowledge . . . moving from the bottom upward." Too frequently they are based 
on "the need for organized procedure in the adjustment of differences," not only 
differences between the workers and the management, but between all kinds of "inter- 
ests," specialized and otherwise. Their time-wasting capacities are notorious. In any 
event there is no reason why many of their functions should not be performed by 
individuals. An integrated system of shop stewards and a chief shop steward in con- 
tact with a Personnel Manager can provide such a service of information upwards in 
an industrial concern. The graduated "staff" positions in the Army are intended to 
do so as far as they extend, and do in fact do so. 

More serious is the petrifaction of leadership which follows from an overload of 
administrative work. A British general has remarked that the huge armies of the great 
war "added vastly to administrative difficulties" and so "paralyzed generalship": 

"As the general became more and more bound to his office, and, consequently, divorced 
from his men, he relied for contact not upon the personal factor, but upon the mechanical 
telegraph and telephone . . . nothing was more dreadful to witness than a chain of men 
starting with a battalion commander and ending with an army commander sitting in tele- 
phone boxes, improvised or actual, talking, talking, talking, in place of leading, leading, 
leading." 42 

But this "disease of generalship," indicated by a soldier who understood "the immense 
value of a staff," if it is properly used, "if it is the general's servant, and not the gener- 
al's gaoler," is endemic in civil life where there are no staffs at all. It is scarcely possible 
to meet a politician or a business man in a position of serious responsibility who is not 
overwhelmed with everyday work. He has no time to read, little time to meet those 
whom he administers. But the most constructive aspects of leadership are those which 
are exercised "face to face." The loss of the "personal touch" which is so often de- 
plored is not due to the size or complexity of modern enterprises. It is due to the 
absence of "staff" organization. Administrators cannot see foremen and fields and 
factories for filing cabinets. 

22. "Staff" Integrates "Line" and Function 

If, as has been suggested, the confusion as to principles of organization in the 
world today is due to an attempt to compromise between the necessity for retaining 
"the scalar process" and the requirements of specialization emphasized by Taylor, 
it will follow that the "staff" conception, if it is a solution, must integrate these 
two essentials of organization. This is so. The "staff" concept in military practice as 
herein described, except in so far as it is not carried below the Company, accords 
very closely with Taylor's ideas. It is avowedly based on his well-known "exception 
principle." It secures that, in so far as is compatible with effective co-ordination, 

42 Major Gen. J. F. C. Fuller, "Generalship — Its Diseases and Their Cure." 


one man is confined to one function: the "staff" itself, in the British Army, is special- 
ized into four branches. It is a clear case of the separation of planning from perform- 
ance. It accords with the irony which sometimes illumines human affairs that the 
difficulties of co-ordination caused in part by the genius of the man who wrote 
"throughout the whole field of management the military type of organization should 
be abandoned," may be solved by reference to the forms of organization which ex- 
perience has forced upon the armies of the world. 

23. Some Objections 

In conclusion certain objections to the considerations which have been put for- 
ward may be considered. The prejudice against militarism which is so widespread 
among Anglo-Saxon peoples, will no doubt be encountered. It will be urged that 
the army is a "disciplined" force and that what is possible in the army is impossible 
or undesirable elsewhere. It should be considered whether in institutions where a 
high standard of discipline is not enforced the problem of co-ordination and the 
provision of mechanism to secure it, is likely to be less serious than in the case of 

The "practical" difficulties of applying technical principles in organization, 
even when their validity is admitted intellectually, are great. Personal factors ob- 
trude. They cannot be ignored. But that they should always and on all occasions be 
given priority in consideration is fantastic. The idea that organizations should be 
built up round and adjusted to individual idiosyncrasies, rather than that individuals 
should be adapted to the requirements of sound principles of organization, is as foolish 
as attempting to design an engine to accord with the whimsies of one's maiden aunt 
rather than with the laws of mechanical science. Individuals are the raw material of 
organization. There is some freedom of choice, some elasticity for adjustment, in large 
enterprises a great deal. That an engineer in the wilderness should build a bridge of 
such materials as he can find is inevitable. That an engineer in a complex industrial- 
ized area should do the same is unthinkable. Even in the wilderness the trained 
engineer will build his bridge in accordance with principles. It will be a better bridge: 
there will be less strain and the materials will last longer. 

Insistence on the personal standpoint in organization almost invariably implies 
an attempt to secure for some individual or for some group special privileges at the 
expense of their colleagues. It may be the chief who prefers to occupy his time with 
detail and his opportunities for straightening out personal situations with patronage. 
It may be subordinate officials who suspect that conservative insistence on the "status 
quo," and not their personal qualifications for positions of responsibility, is the best 
guarantee they have of whatever authority they at present enjoy. It may be elements 
in the rank and file who have developed "defence mechanisms" against some of the 
worst consequences of bad organization, and prefer to maintain their habits rather 
than to risk a change. 

Whatever the motive underlying persistence in bad structure it is always more 
hurtful to the greatest number than good structure. It opens the way to every type 
of dishonesty and intrigue. Such things bring petty miseries every day and larger 
injustices very often. As Mooney and Reiley have observed: 


"How often do we hear of business institutions that their organizations are all 'shot through' 
with politics. . . . Such conditions . . . are really due to inattention on the part of the 
management to the necessities of formal organization, and the application of its principles. 
. . . The . . . type of management, which regards the exact definition of every job and ev- 
ery function, in its relation to other jobs and functions, as of first importance, may some- 
times appear excessively formalistic, but in its results it is justified by all business experi- 
ence." 43 

More serious is the inertia of current practice. Men accept existing forms of 
organization at the point to which they have evolved as a matter of course. That 
they have been different in the past is ignored. The suggestion that enterprises for 
other purposes may have the same needs, be suffering from the same conditions, 
which have compelled that evolution, alone rouses resistance. The resistance is rein- 
forced by what the French call "deformation professionelle," the tendency to look at 
all questions from a particular point of view, to assume that in respect of the indi- 
vidual's own profession what is must be best, which seems to result inevitably from 
the long and intensive practice of a particular calling. It is intellectually incapable of 
imagining alternative arrangements and insensitive to considerations other than those 
current in the given group. It is often associated with great disinterestedness and 
sense of public service. 

24. The Evolution of Organization 

From both these standpoints the evolution of the higher administration of the 
British Army is instructive. Today it is co-ordinated in a single department, the War 
Office. Everyone regards this piece of organization as a matter of course. Naturally 
an army must have one head, must be co-ordinated. In the year of Waterloo a British 
commander in the field had to deal with more than a dozen independent and separate 
ministers or departments in Whitehall on the affairs of his army. It required a hun- 
dred years of struggle, of incessant royal committees and commissions, of recurrent 
scandals due to maladministration, before the last vestige of conflicting authorities 
had disappeared and the constitutional and administrative position was finally 
consolidated. Certain significant stages in this evolution are illustrated in Figure XII. 

At every stage the most eminent generals, the men who had suffered themselves 
under the worst effects of this regime in the field, fought reform — Wellington, 
Wolseley, perhaps the most extraordinary case of all, Airey who had been Quarter- 
master General in the Crimea and had seen the men perishing by hundreds for lack 
of the ordinary elements of workmanlike arrangements. In every case it was deter- 
mined civilians, such as Esher, Cardwell and Haldane, who won the next step. 

Today, the State is taking an increasing part in the administration of economic 
life. There are more than a dozen separate departments and offices in Whitehall 
each dealing with some particular aspect of the subject. The individual business man, 
the general in the field, has to co-ordinate their various services at the point of action 
somehow or other. Everyone regards it as perfectly natural. New semi-independent 
authorities are being created every year. Any suggestion of change of direction meets 
the insuperable objection "The Civil Service wouldn't like it." 

43 Op. cit. 



















b O 



■rf ** K 



(i i. o e 

■a A K 

B O »B 
•0— KB 

□ o o 




B»° . 


=-- B* 

H - 

H — 

-i — ' 


H r 




• B 

O I.H 

• « 

>» c*> 

e -* B 

b a) a 

to BO 

B B B 

Bf B ■ 

I I l" 

a u a o iH»ei(tr 

* a -<-j c o i- ■■_ -it' r ii o 

■a o. i. a a kouceac^ 

b • a w ■ o — a c o-o 

e c o o - -t s , i : a t - 

■ m — iJ— (uChriOjn 

i a>*> *j a c e - o H El- 
's i- i- «-- o n «■-■ .■ — 
: tj o o >j - n u^ri^iC 

> e o. — tdotuii* 

■ — • O " c c 10 « -. -. O 

1 I u u e — c c S >,«. 3 o c 

- B B " L. . I U -. C a ; I , • 

> a c-* ca<oc«ac< 

> «C( O CO.* . . •- i. 



« V. «4 M >,0 

™. 5^ • B~t D *>*• c c tJ-*tt 

b c i. o a ** a o*j« (i i. a> 5> «js 

*J I O Q- O „ tt SO.C4D 0"** 

** t 5 * ** b^vohooobobou* 

OtfO ., *> a o C H " • 

>.k-wu *. b— a ao.*.. 

boo fi n-o >. a a as o t 

J u 

e-Oi U *<ce n li n. < ir. f. v i^ r h 

O— b c b a B • 

T3 B « 3> l~ n IB 
,-< C -> W 4J e^t c 

ii 1 1 


I»hZ<EUU M»ft.n. JH 


c • k *• B —«■ 

o — «i-> o»*-4jc 

«* M . o », cu*cue«ouc:cBeo.c 

■n ■" (L o B <" •? *- *- *• O — u^tlWHU «uomo»c 

- ceo •«• c»o ** o >. obo *- — i rj c o « >>(."< t-. ooo 

o r> " r, l. t/j c «u o t. t. o *• tot. -. •- «f> « >. >. u >. u — 

• tn e « oBbBoi-BBbobao*' •«c->.*'RooBi-i-o--'t--. c 

3 *J C »- « — B C tcu-friOf tUfilJO «B«BBOr"'a>03**oa 

i t! e o ». o -■ 3 o < gu&co'»uui]v(b. .i c-'-'De-' u- •>■ d-h 

*>C «i«*00 >.b«*4->iyjec0OU- « « — 3 i. ax »H &u«« O. 

>nt -chco e o i- 3 »• •- *- i- o i — ■ o *■ i- t? -< -■ a eat.— eawco 

t a o • o « u > — i/i (-. -. * xi — o — r ■ c c a> U — n 3 e^xECEiXi) 

i u <di.i m a m t. v. o a a a «-• coe<o •« a •-■ o o 

; Z!2AJz-l 51i_l_LJ— I -° ■ " I ' ' ■ ' 

J N - 

heu «*b: ou. i*-hji« 

r- ft 7 -- ) ; ( 

» a 

naaBBBb -■sk.aoB 
kcokocce coeofab 

;_t_£_s ° TVV? 

C OO* *• O0-'*'-*«B«BMCOC tOO* 

1> C«"H1. H|lu«3 .fl'o « -< »• C 

U — i. O O O X- J) BO£B CBEBIB ■• »■ I- 

** a txrf e«> >.ao-**-sou' oeei<t*jl*>> 
eu-i i. c r. h i. u .. ,. in a •wnaae^uwuii 

t — u !.': * Q IW 0---1. 1-iL u U C lie I" I _. ti J 

■" C C -J '.1 =N CI- S> *J I. O 1. C O " Hi " CI —<■-» 

3 t. aa d i- o •■ si c a. u n u s -. -■ - 

-:oie co-. c o v- c j3 p » o OB i : n i 


1 -.^-— ,.ff,—- v^/4 

:i ! : 

-\-—> a 

t-mw K<(C r. i>. i*. . -. u w 

)io io *• a*>-ja* a a. v. br -■ c — • 

* o obbb -*je 3Beaa* »»•*».-* *»r< ■ 

I kRolhu^CliOHIil OO VJ -*■— C O Nar4ebRMBB€ 

a ««n.O«»«"tHC D<i4r|t>r4>l u- O « 6 • O c- C i 

O-.- K *> 

CD(- CD fl 

ov a *j b 

go a h>H 

o a** o s *• 

io-^mu « o c 
v>B e • o 

«K >^ 
L.> e c CU.O 


OO I — b*i 

n o 

O fl CM 

so oh a 

Io- a-* 

o-i B>e •* 

oi- s b a 

** o co 

c^o gu « 

-*3 St-A ■-< 

HO 0< B B 

una; U m 

• f a «h«>- 

*>oBfc.3c ao-«-o— ooo> 

CI it «o (I> *> t f l- .; t, u IB — 
t*r4 U >,fll.i itti- Ola U 

4jQ.Baisc ca to oa 
c Bfhov..* i) ceo a 

a Bt3 a •-— 

: a o o t 
dob a c 




i— i 




















■< to 


a *• c • 
oc o" e 

uc — -* o 
*-»- t. c >. 

O O 3 ►• 


H O 4i 

fc.T) 0«»-< ••* 

■< a c o a — 

v- o a ox: o c 
ti a o*> 0,0 -* o 
o o h 1 — 

h a 
h o 

e — < 

<0f o 

o- 5- a* •-* c 


Herbert Spencer once wrote "socially as well as individually, organization is 
indispensable to growth; beyond a certain point there cannot be further growth 
without further organization." Rapid growth in scientific knowledge has placed 
an unprecedented strain on man's powers of organization. The effects of that strain 
are just becoming apparent. Knowledge of and interest in the technique of organiza- 
tion and in its basic principle of co-ordination, are as yet feeble. The common man 
fears the concentration of authority, lest he be unable to control it. The future of the 
Western nations hinges on their power to awaken that interest and to remove that 
fear. "Divide and rule" may be a sound motto for an agricultural despotism. For a 
machine-using democracy it is a passport to disaster. 

"But history records more frequent and more spectacular instances of the triumph of im- 
becile institutions over life and culture, than of peoples who have by the force of instinctive 
insight saved themselves alive out of a desperately precarious institutional situation, such, 
for instance, as now faces the peoples of Christendom." 44 

These words were written in 1914. 

44 Thorstein Veblen, "The Instinct of Workmanship," pp. 24 and 25. 




Vice President, General Motors Corporation 


The word principle signifies something fundamental. So also, in its human aspect, 
does the word organization. These two fundamentals, and how they relate, constitute 
the theme of this study. 

Concerning the fundamentality of human organization, as such, a few words will 

Most of us think of organizations as essentially something big; something that 
unites large numbers of people in some common purpose. Such combinations cer- 
tainly do give high visibility to certain forms of human association, but the word or- 
ganization itself means something far more basic. 

The term organization, and the principles that govern it, are inherent in every 
form of concerted human effort, even where there are no more than two people in- 
volved. For example take two men who combine their efforts to lift and move a stone 
that is too heavy to be moved by one. In the fact of this combination of effort we have 
the reality of human organization for a given purpose. Likewise in the procedure nec- 
essary to this end we find the fundamental principles of organization. To begin with, 
the two lifters must lift in unison. Without this combination of effort the result would 
be futile. Here we have co-ordination, the first principle of organization. Likewise one of 
these two must give the signal "heave ho !" or its equivalent, to the other, thus illus- 
trating the principle of leadership or command. Again the other may have a suggestion 
to make to the leader in the matter of procedure, which involves the vital staff princi- 
ple of advice or counsel. And so on. Thus in every form of concerted effort principles 
of organization are as essential and inevitable as organization itself. 

My own principal interest lies in the sphere of industrial organization, which of 
all major forms of human organization is, in its present magnitude, the most modern. 
For this the reason is evident. The vast present-day units of industrial organization 
are products mainly of one creating factor, namely the technology of mass production, 
and this technology, born of the industrial revolution, has been almost exclusively an 
evolution of the last century. In contrast other major forms of human organization — 
the state, the church, the army — are as old as human history itself. Yet if we examine 
the structure of these forms of organization we shall find that, however diverse their 
purposes, the underlying principles of organization are ever the same. 

On the surface it might appear otherwise. For example we may think of line and 
staff as something peculiar to military organization — and so they are if we consider 
the origin of the terms only, and the particular functions to which they refer in mili- 
tary procedure. But if we take the broader view and consider line simply as the func- 
tion of command, and staff as the service of information or counsel, we see at once that 
military organization can have no corner on these principles. The organizations of 
civil government have and of necessity must have the equivalent of these functions, so 
has church government, and so likewise has industrial organization. 

The point here is that a principle, if it be truly such, must of necessity be uni- 
versal in its own sphere, and so it is with the principles of organization. There is no 



principle in industrial organization as such that is not to be found in all of these other 
spheres, but it is erroneous to infer that industrial organizers have borrowed these 
principles of organization from these older forms. A principle, if it is truly such, is a 
universal, and a universal cannot be borrowed. It simply has a way of applying itself, 
and this is ever true, by whatever name we may be pleased to call it. If we can but 
recognize a principle when we see it, it is a simple matter to identify and co-ordinate 
what, for present purposes, I have called the principles of industrial organization. 

Organization, as I have said, means concerted human effort, the kind of effort 
that is essential to the highest measure of success of any group undertaking. But what, 
in industrial organization, are the conditions necessary to the highest efficiency of 
such concerted effort? If we addressed this question to a score of representative in- 
dustrial executives, we would of course get many valuable and inspirational answers. 
But I wonder how many of these answers would give the structural principles of or- 
ganization the importance that I believe they deserve? 

It may be asked why I emphasize the term "structural." The reason is that many 
people think of organization as in some way synonymous with administration or man- 
agement. In every organization there is a collective job to be done, consisting always 
of the sum of many individual jobs, and the task of administration, operating through 
management, is the co-ordination of all the human effort necessary to this end. Such 
co-ordination, however, always presupposes the jobs to be co-ordinated. The job as 
such is therefore antecedent to the man on the job, and the sound co-ordination of 
these jobs, considered simply as jobs, must be the first and necessary condition in the 
effective co-ordination of the human factor. 

The importance of the exact functional definition of each job as such, in its effect 
on the morale and general efficiency of an organization, may be shown by illustrations 
that are familiar in all business experience. How often do we hear it said of business 
institutions that their organizations are all "shot through" with politics. A superficial 
thinker might take this as a reflection on the personnel. If he should become ac- 
quainted with this personnel he might be surprised to find how good, potentially at 
least, it really is, comparing not unfavorably with the personnel of other institutions 
that function smoothly and harmoniously. Ten to one we must go to organization, 
rather than personnel, to find the real cause of the trouble. 

Such conditions are really due to the inattention on the part of administration to 
the necessities of formal organization, and the application of its principles. When an 
employee is placed in a position, with duties ill defined in their relation to other duties, 
what happens? Naturally he attempts to make his own definition of these duties, and, 
where he can, to impose this view on those about him. In this process he encounters 
others in similar case, with friction and lack of co-ordinated effort as the inevitable 

On the positive side orderly procedure gives way to the practice of "cutting cross 
lots"; on the negative side it results in the shirking of responsibilities, or, in popular 
phrase, "passing the buck." Such conditions become aggravated when management 
itself begins to take short cuts, without consideration of the longtime consequences. 
The two conditions usually go together, for the management that is inattentive to the 
definition of subordinate functions is almost sure to be just as disorderly in the exercise 


of its own. True co-ordination in the formal sense can only be effectuated through 
functional definition, and such co-ordination must begin at the top. Without it there 
will be friction, even at the top, and in these circumstances it is futile to look for co- 
ordinated harmony at any other point down the line. No one who has ever had 
personal knowledge of such conditions can question the importance of the exact 
functional definition of all jobs as such, in their relation to other jobs, in its effect on 
collective efficiency and collective morale. 

The other type of management, which regards the exact definition of every job 
and every function, in its relation to other jobs and functions, as of first importance, 
may sometimes appear excessively formalistic, but in its results it is justified by all 
practical experience. It is in fact a necessary condition of true efficiency in all forms 
of collective and organized human effort. 

It is such co-ordination that I have in mind when I speak of the structural princi- 
ples of organization. A good job of organizing, in this sense, is a necessary condition of 
any efficient and successful management. I shall now endeavor to describe, as briefly 
as I can, those structural principles of organization, the application of which is neces- 
sary to this end. 

Co-ordination, as we have noted, is the determining principle of organization, 
the form which contains all other principles, the beginning and the end of all or- 
ganized effort. We must find the actual process, however, through which co-ordina- 
tion becomes effective throughout an organization. Here I would distinguish between 
two forms of co-ordination, the perpendicular and the horizontal. I cannot conceive of a 
truly efficient organization of any kind that does not illustrate the operation of both of 
these principles. 

The principle of perpendicular co-ordination is expressed in the single word au- 
thority. By authority I do not necessarily mean autocracy. In democratic forms of or- 
ganization the supreme authority may be represented in the group as a whole, as it is 
under our government in the people of the United States. Nor is authority something 
that, under any system, can ever be segregated at the top. Responsibility without 
corresponding authority is inconceivable, and sound organization demands a clearly 
defined responsibility for every act, from the greatest to the smallest. 

It follows, therefore, that authority must have a clearly defined process through 
which it projects itself throughout an entire organization, so that everyone in the insti- 
tution participates in the exercise of this authority, according to the nature of his duties. 

Here we come to what I conceive to be a vital distinction; that between authority 
as such, and the form of authority that projects itself through leadership. The difference 
may be seen in their relation to the organization itself. It takes supreme co-ordinating 
authority to create an organization; leadership, on the other hand, always presupposes 
the organization. I would define leadership as the form in organization through which 
authority enters into process; which means, of course, that there must be leadership as 
the necessary directive of the entire organized movement. 

We know how leadership functions in the direction of this movement, and we are 
all familiar with the structural form through which it operates. We call it delegation of 
duties, but few realize how absolutely necessary to an orderly and efficient procedure 
is a sound application of this delegating principle. 


Delegation is not a transfer of authority; it is a correlation of authority, and like- 
wise of responsibility. The one to whom a task or a job is delegated becomes responsi- 
ble for doing that job, but the superior who delegates this authority remains respon- 
sible for getting the job done. And this chain of correlated responsibility, which I call 
the scalar chain, extends from the top to the bottom of the entire organization, but al- 
ways emanating from the top leadership, which is responsible for the whole. 

Delegation, as a principle, is universal in organization, and no form of concerted 
human effort can be conceived without it. It begins where organization begins, in the 
simplest relations of superior and subordinate, and, no matter how vast the organiza- 
tion becomes, or how much the scalar chain may lengthen through sub-delegation, the 
principle of correlated responsibility remains ever the same. 

But this process of delegation cannot and must not be conceived as an end in it- 
self. It must have its own aim and end; in other words its own final purpose. This pur- 
pose is nothing less than the co-ordination of functions . 

By functions I mean distinctions between kinds of duties. We know that authority is 
the determining principle of organization, but it is impossible to think of the variations 
of different jobs in terms of authority alone. Functionalism enters from the very first; 
from the first the central line of authority begins to throw off functions, and in the end 
it always breaks up into functional distinctions. Authority, represented in leadership, 
and operating through delegation of duties, has only one aim and purpose within the 
organization, namely the co-ordination of functions. And the efficiency of such co- 
ordination is the measure of the efficiency of the organization itself. 

This is certainly true of the executive group. It is no less true of the organization 
as a whole, and equally true of every organization, whatever its purpose. We see this 
illustrated in a well organized factory, geared to mass production, with its countless 
individual jobs or functions, many of them seemingly small in themselves, but all of 
them fitting into the general scheme, and essential to the collective result. We see it 
again in another sphere in the perfect co-ordination of a symphony orchestra, where 
the co-operation of every individual musician is likewise essential to the collective 

We must not think of functionalism, however, as something that always has pri- 
mary reference to the function of the individual as such. Departmental functionalism, if 
I may call it by that name, is also a vital factor in nearly all forms of organization. 

Take for example the military. The distinction between generals, colonels, ma- 
jors, captains, etc., down to the high privates, is clearly scalar, and always defines it- 
self as such. But just as clearly the distinction between the infantry and the artillery is 
functional, and the same applies to every other special function in military organiza- 
tion, including the auxiliary functions of supply and transport. 

This same phenomenon of departmental functionalism appears in every sphere of 
industrial organization. It may be less evident in the sphere of production, where 
functionalism concerns itself mainly with individual duties. In the sphere of distribu- 
tion, however, no matter what the nature of the product, there are four divisions of 
departmental functionalism that appear to be well nigh universal. These four are 
finance, sales, supply, and service. 

If any of these four can be identified as the principal or determining function, it is 


obviously finance, and so it appears in most forms of distributive organization. The 
other three distributive functions appear in all kinds of relationships, according to the 
nature of the industry. In some industries supply and service are subordinated to sell- 
ing in functional importance. In others this relation is reversed, and the two in com- 
bination appear as a far more important functional arm. The latter is generally true 
of all public utilities. 

From the standpoint of organization, departmental functions of such importance 
require scalar organization from the top downward, such functions, at each stage of 
the scalar chain, adhering to the corresponding line authority. In this respect they 
clearly correspond to the line and staff functions in army organization, a subject on 
which I shall presently have more to say. The point I here wish to emphasize is that 
the organization of any function of such importance from the very top downward is an 
absolute essential to any truly efficient co-ordination of effort. 

The necessity for such co-ordination, in order to attain a true efficiency of collec- 
tive effort, is too obvious to require further illustration, but how to attain such 
efficiency carries us back to the process that assigns all functions. The scalar chain in 
organization, with its authority and delegation of duties, is the source of this co- 
ordination; it is here, in the way these principles are applied, that we may find the 
source either of a collective harmony or a functional discord. 

So much for perpendicular co-ordination, operating through leadership and the 
delegation of authority, without which no organization can function. Another factor 
of equal importance is that of indoctrination in the common purpose, which is obviously 
essential to the true intelligence of concerted effort. Here enters the other great prin- 
ciple, that of horizontal co-ordination, which operates not through authority and the 
function of command but through the universal service of knowledge. This difference 
between the perpendicular and the horizontal forms of co-ordination brings us to the 
final distinction in functionalism, that between the line and the staff. 

As students of organization we have become familiar with the military terms line 
and staff, and we have witnessed the extension of these terms to the sphere of indus- 
trial organization. But we are still prone to think of the line and staff distinction as in 
some way an evolution out of military forms. 

My study of human organization refutes this notion. It is true that the terms line 
and staff are military in their origin, but that is all. The principles they describe are as 
old as human organization itself. In one form or another, and under various names, 
this distinction appears in other forms of organization — of church or secular govern- 
ment — from the beginnings of recorded history. In every form of organization we 
find the function of authority or command, and likewise the function of advice or 
counsel, the latter constituting a true service of knowledge. Thus the universality of 
this distinction, the true test of a principle, is clearly established. It is this same dis- 
tinction that I have in mind when I contrast the principles of perpendicular and of 
horizontal co-ordination. 

Authority, with its process of delegation, is of course essential in every organiza- 
tion, but experience proves that the service of knowledge is an equal necessity, and, 
furthermore, that such service is impossible through the contacts of command alone. 
Staff services, whether formally organized or not, are bound to grow up in every or- 


ganization. Their formal organization, however, is demanded if we are to achieve the 
most efficient forms of concerted human effort. 

Staff service, like line authority, begins at the top. In this phase it is informative 
and advisory to the top leadership. It has been defined as an extension of the personal- 
ity of the executive — more ears, more eyes, and more hands to aid him in the formu- 
lation of his policies. These however are the informative and advisory phases of staff 
service. They become supervisory when all the formed policies of leadership are trans- 
lated into execution. It is in the latter phase that we find how the service of knowledge 
is infused throughout an entire organization. 

Here military organization furnishes an impressive example. We notice that 
military staff organization is scalar, like the line to which it adheres, following at each 
step in the scale the gradations of line authority. Even the smallest units of command 
have their own staff services. But more than this is needed if an entire organization is 
to be permeated with a true service of knowledge. 

It is not the leader alone who has things to make known to his subordinates 
through the usual channels of staff service. These subordinates may likewise have 
something important to tell the leader; things that he should know in the exercise of 
his leadership. They may also have important things to tell each other, and this mu- 
tuality of things to be made known extends upwards, downwards, and sideways, from 
the very top to the bottom of the organized structure. 

If we look about us in the sphere of industrial organization we may see examples 
of how such a service may be organized, prominent among them the interlocking com- 
mittee systems. No intelligent movement toward a common objective is possible with- 
out such a service. If military organization, with its greater intensity of discipline, 
finds the service of knowledge so imperative, how much more evident the same need 
becomes in the case of every industrial objective. In all forms of organization, what I 
have called horizontal co-ordination is the principle that indoctrinates every member of 
the group in the common purpose, and thus insures the highest collective efficiency 
and intelligence in the pursuit of the objective. 

In this outline of the structural principles of organization I have done no more 
than expose the scaffolding, the frame-work out of which it is made. In stressing the 
importance of this frame-work, I hope no one will think that I am overlooking the 
major importance of the human factor. Humanly speaking, the strength of any or- 
ganization is simply the aggregate strength of the individuals who compose it. We 
cannot forget, however, that the strength of the individual, whatever that strength 
may be, can only attain the highest measure of effectiveness through soundly ad- 
justed relationships, and it is here that we see the fundamental importance of these 
structural principles. If we truly co-ordinate the jobs as such, we shall find that the 
more efficient and harmonious co-ordination of the people on the jobs is immensely 

One question frequently asked is whether the development of personnel policies 
is properly a line or a staff function. The obvious answer is that it is both, for there 
can be no staff service in the segregated sense. Such service must always adhere to line 

Nevertheless, in the development of personnel policies, the staff should be a po- 


tent factor. It is a point worth observing that in the general staff of our army the first 
division, or G-l, is the personnel division. This division, furthermore, has nothing to do 
with operations and training, which is the concern of a different section. It is con- 
cerned with personnel simply as individuals, and it advises concerning the movements 
of men in order that the right man shall always be found in the right job. Of course on 
the human side there is more to personnel policies than the mere fitting of the man to 
the job. Nevertheless this is always the prime essential — the one that must have first 
consideration in all personnel plans. 

Finally, in all of this discussion concerning principles of organization, and how 
the application of these principles contributes to organized efficiency, we must keep 
ever in mind the fact that the prime importance of industrial organization, like every 
other form of human association, lies in its aim or object, and this leads us to the con- 
sideration of how the other great factors in human affairs, governmental and social, 
may affect its policies, aims, and even its destinies. Granted that efficiency is essential 
in everything that is worth the doing, we must not forget that efficiency is never an 
end in itself. It must always have an aim and purpose and it can only justify itself 
through the worthiness of this purpose. Worthiness in the industrial sphere can have 
reference to one thing only, namely the contribution of industry to the sum total of 
human welfare. On this basis only must industry and all its works finally be judged. 

Even from the standpoint of efficiency in the pursuit of its objectives the record 
of industry during the last forty years leaves something to be desired. The growing 
efficiencies of industrial production during this period are familiar and we know 
what these efficiencies have accomplished in the lowering of production costs. What 
we also observe, however, during this same period, is that these producing efficiencies 
have not always had the effect they should have had in the reduction of price to the 
final consumer. And this fact brings us face to face with something that we all should 
know, namely that distribution, and the rising costs of distribution, and not produc- 
tion, constitute today the great unsolved industrial problem. 

During this period the increasing efficiencies of production have created a de- 
pendence on ever larger and wider markets, and we have not developed the corre- 
sponding efficiencies in distribution that these conditions compel. The efficient pro- 
duction engineer has long been with us, but the distribution engineer is a more recent 
arrival, and in the proper relating of plants and products to markets he still has his 
own major problems to solve. 

It is not such problems, however, but rather the moral problems growing out 
of the human relations in industrial organization that constitute our main concern 
at present. 

Here we encounter the present-day problem of industry in its most vital aspect. 
Organizations, after all, are composed of people, and no organization can be any 
stronger than the people who compose it. To be exact on this point, the great in- 
ternal problem of industry on the human side is and always has been the relation of 
the individual to the group, and the further fact that the reasonable continuity of 
this relation is one of the prime essentials of group loyalty. This brings us face to face 
with the problem of social security, now uppermost in the popular mind. Four and 
a half years ago, -in June 1932 to be exact, in an address on "Current Problems of 


Industrial Management," I made the prediction that if industry does not step up to 
this problem government will. We all know how that prediction has been fulfilled. 

It is obvious that these later developments, which have made the government 
the sponsor for social security, do not help to improve the strength of industrial 
organization, for under present policies the security, if it indeed be such, appears to 
come from elsewhere. The fact remains, however, that industry, which always pays 
for everything, must finally pay the bills. What industry must continue to strive for 
are measures that will help to insure the continuity of the human relationship, with- 
out which no organization, however potent it may appear at any given time, can 
be assured of enduring strength. There is no escape from the fact that continuity of 
employment is the sole basis of enduring group loyalty, and this continuity, as all 
experience proves, is in turn dependent on reasonably stable industrial conditions. 
Nothing, as all experience likewise proves, can give absolute assurance of such con- 
ditions, but a much nearer approach is feasible through a scientific study of markets, 
and a sound relating of plants and products to markets, which should be the principal 
care of industrial engineering, now and hereafter. 

In conclusion I would emphasize the fact that, despite these considerations, 
organization, and the sound application of its principles, must always remain a potent 
force in determining industrial destiny. Organization concerns procedure, and the 
attainment of any human group objective must ever depend, in great measure, on 
efficient forms of procedure. The lessons of history teach us that no efficiency of 
procedure will save from ultimate extinction those organizations that pursue a false 
objective; on the other hand, without such efficient procedure, all human group 
effort becomes relatively futile. It is in these facts, I believe, that we may see clearly 
the importance of these principles of organization, in their bearing on future in- 
dustrial progress. 




Translated from the French 

Sarah Greer 

Institute of Public Administration 




In his address at the opening of the Congress of 1910 the President, M. Cooreman, 
gave the following definition of administration : 

"By administration we mean three things connected but distinct: an ensemble of the 
executive power and the exercise of this power; the body of functionaries and employees; and 
the administrative personnel. 

"The executive power as we understand it here, has for its object the maintenance of 
order, public security, the guarantee of a just and free use of common property, the manage- 
ment of the public wealth, the execution of the laws and the preparation and putting into 
effect of measures of general interest." 

According to this definition the science of administration would include only 
knowledge relative to the services, agencies, the personnel and the operation of public 
administration. In fact the Congress of 1910 had in view only the administration of 
public affairs. 

The meaning that I have given to the word administration and which has been 
generally adopted, broadens considerably the field of administrative science. It em- 
braces not only the public service but enterprises of every size and description, of every 
form and every purpose. All undertakings require planning, organization, command, 
co-ordination and control, and in order to function properly, all must observe the 
same general principles. We are no longer confronted with several administrative sci- 
ences but with one alone, which can be applied equally well to public and to private 
affairs and whose principal elements are today summarized in what we term the Ad- 
ministrative Theory. 

This seems to be also the opinion of those who planned this conference, for in 
their invitation we find the following: 

"The importance of administration has grown steadily since the first Congress of Ad- 
ministrative Science held in Brussels in 1910. In consequence we have felt strongly the need 
of good administrative methods, and men such as Fayol, Solvay and Taylor have in recent 
years worked out certain formulas (or a synthesis of principles) which should in their opinion 
govern an administration. 

"Belgium has made many successful experiments in this vast field. The Ministry of Na- 
tional Defense has just been reorganized according to the principles of Fayol. Some impor- 
tant steps in this direction have also been taken by the Ministry of Agriculture." 

Following the example of Belgium, France has recently been inspired to reor- 
ganize the Department of Posts, Telegraphs and Telephones according to the Adminis- 
trative Theory. In an act presented to the Parliament in July, 1 922, the Government 
said as follows: 

"Already many government activities in foreign countries have put into operation the 
administrative methods that we wish to introduce into the administration of the Postal De- 
partment. In Belgium the Minister of National Defense has published a brochure which con- 
tains an outline of the ideas of M. Fayol which have been successfully put into operation in 
his department. 

"France must not be the last country to apply modern methods to public administra- 



We must not be deluded, however, by these official statements for in reality, the 
essential principles of the Administrative Theory do not seem to have been applied in 
their entirety. Even the organization and operation of industrial undertakings sel- 
dom conform to them, while governments are only waking up to the theory of ad- 
ministration and so far have limited themselves to certain questions of detail, with- 
out attacking the real problem. 

The essential principle of the Administrative Theory is the great importance of 
management and this importance increases with the size of the undertaking. No enter- 
prise can be successful without good management and every enterprise which is 
poorly managed is doomed to failure. This principle is as true for the State as for 
private industry. In my opinion, if the operation of the public service causes so much 
vexation it is because it is not well managed. The manner in which the subordinates 
do their work has incontestably a great effect upon the ultimate result, but the opera- 
tion of the management has a much greater effect. The Congress of 1910 was con- 
cerned only with the work of the subordinate governmental employees, ignoring the 
machinery which is of first importance in the operation of any business. It is of this 
machinery that I am going to speak today. 

I will first take up, from the point of view of the Administrative Theory, what 
should be the role, the means of operation and the structure of the high command 
in the public service, then I will glance at the manner in which these problems are 
understood and solved. I shall take up here only the question of agencies and shall 
not touch upon personnel. I assume that all the members of the Congress are familiar 
with the Administrative Theory. A resume of it will be found at the end of the printed 
report of my address. 1 

Role, Means of Operation and Structure of the High Command in the 

Public Service 

The structure of the high command in the public service has the same general 
aspect in all modern states; under different names we find almost everywhere a Prime 
Minister, ministers and directors. 

The Prime Minister has authority over the entire governmental enterprise. It is 
his duty to conduct the enterprise towards its objective by endeavoring to make the 
best possible use of the resources at his disposal. He is the head of the ministers and 
must see that all essential functions are carried on. Each minister is in charge of a 
group of activities and is the head of the directors of his group. Each director is in 
charge of a certain activity and is the head of all the employees in his activity. 

According to the worth of the men who occupy these three grades of responsi- 
bility and according to the means of operation at their disposal, this general plan of 
the high command in public administration has results ranging from the best to 
the worst. 


Let us see what is the principal role of the high command in the public service. 
In a great enterprise like the state, this role is essentially administrative; it consists 

1 This resume is not given here, since an outline of Fayol's theory is to be found in L. Urwick's: "The Function of 
Administration with Special Reference to the Work of Henri Fayol," Paper V, pp. 115-130. 


in preparing the operations of the various governmental services, in seeing that they 
are carried out and in watching the results. To prepare the operations is to plan and 
organize; to see that they are carried out is to command and co-ordinate; to watch the 
results is to control. 

The preparation of the operations is the result of a twofold effort of planning and 
organization. To plan is to deduce the probabilities or possibilities of the future from 
a definite and complete knowledge of the past. To organize is to define and set up the 
general structure of the enterprise with reference to its objective, its means of operation 
and its future course as determined by planning; it is to conceive and create the 
structures of all the services that compose it, with reference to the particular task of 
each. It is to give form to the whole and to every detail its place; it is to make the frame 
and to fill it with its destined contents. It is to ensure an exact division of administra- 
tive work by endowing the enterprise with only those activities considered essential 
and with careful determination of the sphere of each of them. Thus in organization, 
the theoretical concepts of planning are translated into facts. 

Execution is the result of command and of co-ordination. To command is to set 
going the services defined by planning and established by organization. 

An order when given, sets into motion simultaneously in all the grades con- 
cerned, authority and responsibility, initiative and discipline. 

But command would be powerless to ensure the complete execution of the will 
of the chief, if it were not supplemented by co-ordination. To co-ordinate is to bring 
harmony and equilibrium into the whole. It is to give to things and to actions their 
proper proportion. It is to adapt the means to the end and to unify disconnected efforts 
and make them homogeneous. It means establishing a close liaison among services 
specialized as to their operations, but having the same general objective. 

Control is the examination of results. To control is to make sure that all operations 
at all times are carried out in accordance with the plan adopted — with the orders 
given and with the principles laid down. Control compares, discusses and criticizes; 
it tends to stimulate planning, to simplify and strengthen organization, to increase the 
efficiency of command and to facilitate co-ordination. 

Such is the administrative role that our theory assigns to the high command in 
the public service, a role in which the Prime Minister, the ministers and the direc- 
tors have each a part corresponding to their functions. 

This role is much the most important one for each of these three upper grades, 
but it is not their only one. Besides his administrative role, the director of a public 
service must see to the carrying on of all important functions in accordance with the 
orders he receives from the minister. He must naturally follow closely the professional 
operations characteristic of his service (financial, military, judicial or industrial), and 
this requires of him an outstanding technical ability in his field. 

The minister receives his directions from the Prime Minister, interprets them and 
transmits them to his immediate subordinates. He controls the actions of these sub- 
ordinates and represents his Ministry before Parliament. 

The Prime Minister, placed at the summit of the administrative hierarchy 
through the confidence of the Chief of State and of Parliament, has charge of the in- 
terests of the country as a whole. He selects the ministers and submits their names to 


the Chief of State and to Parliament for approval; and on important questions gives 
them their directions or indicates the course to be followed. He plans, organizes, 
commands, co-ordinates and controls the public service as a whole. Grave questions 
of general interest constantly demand his attention and take a great deal of his time 
and he must also maintain relations with foreign powers. All of this imposes upon 
him an enormous responsibility which is often beyond the strength of a man most 
capable physically and mentally. 

Such is the role of the chiefs at the three upper levels of public service. 

Means of Operation 

What are the means of operation at their disposal? Let us suppose that the per- 
sonnel of each activity at every rank of the hierarchy are efficient and equal to their 
tasks. Let us also suppose that Directors, Ministers and the Prime Minister are equal 
to their calling. Will this suffice to ensure the efficient operation of the Government? 
Without hesitation I say No. 


Whatever their ability and their capacity for work, the heads of great enterprises 
cannot fulfil alone all their obligations of correspondence, of interviews, of confer- 
ences and of countless other duties; they must ensure command and control, superin- 
tend reports preparatory to decisions, have plans of operations drawn up; encourage 
and effect improvements. Thus they are forced to have recourse to a group of men 
who have the strength, competence and time which the Head may lack. This group 
of men constitutes the Staff of the Management. It is a help, or reinforcement, a sort 
of extension of the manager's personality, to assist him in carrying out his duties. The 
Staff appears as a separate body only in large undertakings and its importance in- 
creases with the importance of the undertaking. The staff of a Prime Minister, of each 
Ministry and of each Direction includes: a secretariat, consultants and accountants. 
The Prime Minister has also a Council for Improvements or Reforms. 

Administrative Tools 3 

Administrative tools are essential to the management of any great public business. 
They are a vast documentation which includes the present, the past and the future, 
to which the elite of the personnel contribute and which, together with its other sources 
of information, enables the management to make under the best possible conditions, 
decisions whose consequences and repercussions can be foreseen. 

It is the practical means by which planning, organization, command, co-ordina- 
tion and control are carried out. It is obvious that in order to manage an undertaking 
the manager must have a thorough understanding of it. He must be conversant with 
all that concerns its objectives, its needs, its resources (raw materials, plant machinery, 
capital, staff, surroundings, etc.), but the acquisition of this knowledge requires a 
great deal of time and effort, and if studies have not already been made and clear 
and complete reports prepared, the manager in office may lack the information that 
he requires and a new manager remain for a long time ignorant of the most essential 
information about the business. 

1 Outillage administratis 


I am going to describe briefly the administrative tools that I used during fifty 
years in a great mining and metallurgical undertaking. Naturally the different parts 
of the machinery were adapted to the nature and importance of each of the units 
which formed the concern, but the general outline of the documentation was always 
the same, and it can be used in every kind of undertaking. It is as follows: 

1 . General Survey. — A general survey (of the present situation as well as of the 
past and the probable future of the business) ; this survey is made for each unit and 
also for the business as a whole. The review of the past history of the concern is made in 
order to recall the reasons which determined the founding of the business, the changes 
it has undergone, and the results obtained. The study of the present situation relates to 
the actual condition of all the parts and of the whole of the resources and needs of 
the undertaking envisaged from every point of view. The probable future is that which is 
foreseen in taking into account the past, the present and any economic, political and 
social changes that may take place. 

In order that this survey may be made efficiently, it is necessary to have an ex- 
perienced head, expert in the management of men, capable of getting a loyal and 
active co-operation from his subordinates and taking a large part of the responsibility 
that the survey implies. From this study should emerge the general scheme and the 
directives which serve as a basis for the Plan of Operations. 

2. Plan of Operations. — The plan of operations is the union or synthesis of 
various plans: annual, long term, short term, and special. It is a sort of picture of the 
future where approaching events are set forth clearly and remote events appear 
vaguely in proportion to their distance. It is the progress of the undertaking foreseen 
and prepared for a certain length of time. The need of a plan of operations is recog- 
nized by all, but the practice is not yet general. Many private undertakings leave 
much to be desired in this respect; while in government, planning is still the excep- 
tion and not the rule, because the preparation of a plan requires a great effort on the 
part of the higher personnel and also a competent and stable head, aided by a good staff. 

3. Reports or Proceedings. — The report on operations carried out is the com- 
plement of the plan of operations. 

Reports of subordinates to their superiors are established for each rank of the 
hierarchy; they are daily, weekly, monthly or annual, and are a powerful method of 
control. The use of the plan of operations and the detailed report by each grade of 
the service permit us to realize two highly important administrative objectives, which 
are the sense of responsibility among employees and confidence among the adminis- 
trative authorities. 

4. Minutes of Conferences between Heads of Departments. — There is a weekly 
conference of heads of offices, bureaus and departments with the manager. Each 
office head gives an account of the work of his department, its accomplishments and 
the difficulties encountered. After the discussion the manager makes his decisions. 
Everyone leaves the conference knowing what he has to do and knowing also that 
he must give an account of it. In one hour the manager has reviewed the principal 
happenings of the past week and the plans for the following week, and this is a 
powerful method of co-ordination and control for him. 


The minutes of this conference, where all the activities of the business are un- 
folded and explained by the leading participants are of the greatest importance to 
the general management. No report or any amount of correspondence could give 
him such a perspective of the personnel. 

5. Organization Chart. — Organization charts, with branches like genealogical 
tables, permit us to seize at a glance, better than we could with a long description, the 
organization as a whole; the various activities and their boundaries; the ranks of the 
hierarchy; the position occupied by each employee, the superior to whom he reports 
and the subordinates under his control. The organization chart draws the attention to 
overlappings and encroachments, to dualities of command and to offices without in- 
cumbents. It is a kind of model which shows the imperfections of the staff as a whole 
and which can be used each time the whole or any part of the organism is reorganized 
or modified. Accompanied by a clear definition of the functions of each, it defines 
their responsibilities and permits us to decide quickly what employee we should apply 
to, in order to deal with a certain matter. 

These administrative tools are indispensable in the management of large enter- 
prises. They permit the carrying out under good conditions, of planning, organization, 
command, co-ordination and control, or in short, efficient administration, which 
without them would not be possible. 

I should like to call attention to two very important administrative results 
which we can accomplish by their use and whose general absence in government 
is one of the chief reasons for the inefficient operation of the public service : the sense of 
responsibility among employees and confidence among administrative authorities. 


A plan of operations for an activity, drawn up by the head of the activity, places 
an obligation upon him, whatever his rank in the hierarchy. Most of the heads 
of activities take part in drawing up the plan of operations. The weekly, monthly or 
annual report, by contrasting plan and accomplishment, shows to what extent these 
obligations have been fulfilled, and from this springs the sense of responsibility. With- 
out a plan there is no obligation, and without a report there is no comparison of 
accomplishment with plan, and thus no responsibility in either case. The absence of 
plans and reports creates a general irresponsibility among employees of the State. 

Confidence Among Administrative Authorities 

In a corporation, confidence among the directors results from the fact that the 
plans of the chairman are known and approved by the board of directors who are 
constantly kept in touch with activities and results, while in the state a minister takes 
charge of a department about which he knows little and which he administers without 
a serious or definite plan. Both the Parliament and the employees are ignorant of his 
intentions, and uneasiness and distrust prevail. But let us suppose a plan of operations 
drawn up by the employees, agreed upon by the minister and submitted by him, first 
to the Cabinet and then to Parliament. There is thus a unity of opinion. The Parlia- 
ment knows what the minister intends to do; regular reports show that the plan is be- 
ing faithfully carried out and confidence takes the place of the distrust which now pre- 
vails among government officials. 


The co-operation that is established among employees in all ranks of the hier- 
archy by the preparation of reports, is another advantage of administrative tools. 
This constitutes a real participation in administration by minor employees and is per- 
haps the best participation that could be devised. 

Administrative tools have many other virtues. They imply that no one will em- 
bark upon any course of action without having foreseen the consequences. This means 
putting planning into operation, and thus an imperfect machine will not be destroyed 
without a better one to replace it. One would not turn over a government undertaking 
to private enterprise, or the reverse, without being quite sure that the change would be 
beneficial to the Nation. By putting these administrative tools into operation in the 
state we might bring about a considerable modification in the orientation of public 

Structure and Organization 

Given the role and means of operation of the high command in the public serv- 
ice, what should be its structure and organization? We will assume that Directors, 
Ministers and Prime Ministers are amply endowed with physical, intellectual and 
moral qualities, as well as with general culture, and will leave these aside to confine 
ourselves to questions of competence and of organization. 


The Director of a governmental activity must have the following qualities and 
competence, necessary to the head of any great private enterprise: administrative 
ability; professional competence appertaining to the enterprise; general notions on all 
essential functions, and stability. If he can keep his post for ten years he is considered 
stable, if he cannot stay in one place longer than a year or two he is unstable. He 
should be supplied with a good staff and good administrative tools. 

The Minister 

The Minister must be a good administrator, able to plan, to organize, to com- 
mand, to co-ordinate and to control. He should have a wide general knowledge of the 
affairs for which he is responsible, but he is not required to have a special competence 
in the various professions characteristic of his department, this being the function of 
the Directors. The importance of stability has not yet been realized by parliamentary 
governments and this is not one of the requirements for a minister. He must, however, 
have platform eloquence, for he plays his principal role in Parliament. There he is the 
representative of the activities under his administration and must ask for necessary 
funds for them and defend his requests against attacks. He is called a good minister if 
he wins the approbation of the deputies and senators, but more than parliamentary 
eloquence is needed to direct a ministry. It is conceivable that a good lawyer might 
become a brilliant minister in a few weeks, but his talent as an orator would not make 
him a good administrator. Finally the minister should be provided with a good staff 
and good administrative tools. His staff should include accountants independent of 
those under the directors. 


Prime Minister 

The Prime Minister should above all things be a good administrator. He cannot 
be expected to have a profound personal knowledge of all the problems incident to the 
functioning of the government enterprise, but usually he has had long political ex- 
perience and with the assistance of a good staff and good administrative tools he can 
make decisions with a thorough knowledge of the case. His staff should include a 
secretariat, special consultants, accountants and a Council for Improvements and Re- 
forms. The Prime Minister must possess parliamentary eloquence to a high degree. 

The Actual State of Affairs 

The above would be the recommended organization of the high command in 
the public service according to the Administrative Theory. It would be interesting to 
see to what extent this idea is grasped and carried out in the various constitutional 
States, but today I must confine myself to a rapid glance at the situation in France, 
feeling convinced that I could apply most of my observations and conclusions to the 
other countries. 

Prime Minister 

In France, for several decades the Prime Minister has been surrounded by from 
twelve to fifteen ministers and several under-secretaries of state. Besides general ad- 
ministrative supervision over the government as a whole, the Prime Minister is in 
charge of one of the government departments and usually one of the most important. 
As Prime Minister he has neither a secretariat, nor accountants, nor special consult- 
ants, nor a Council for Improvements and Reforms. He has besides, no administrative 
tools. He would not even have an office or attendants but for the government depart- 
ment of which he is in charge. 

Aside from his personal worth which is usually high, the Prime Minister has at 
his disposal none of the various means of operation recognized as indispensable if he is 
to fill properly his office as Head of the Government. He has too many ministers 
to direct, and it is absolutely impossible for him to administer the Public Service as 
a whole efficiently and at the same time to be head of a department. There we have 
an extraordinary and incredible fault in organization. Whatever the circumstances 
that brought about this state of affairs there is no reason to maintain it in the face of 
the enormous harm that it causes the Nation. The remedy lies in the suppression of 
portfolios, in the reduction of the number of ministers, in the setting up of a staff, and 
in administrative tools. 


In order to ensure the successful operation of his department the Minister should 
be a good administrator with a certain competence in the affairs of which he has 

He should receive orders from only one chief and have but a small number of 
subordinates directly under him. But in actual fact the Minister is seldom a good ad- 
ministrator who has learned to plan, to organize, to command, to co-ordinate and to 


control, and he has rarely had any training for his position. He is responsible for the 
activities of his department to the Prime Minister and to Parliament, but he renders no 
report to either. If there is any happening out of the ordinary, of such a nature as to 
arouse Parliament or upset the Cabinet, the Minister discloses it and the President 

The Minister usually lacks the administrative ability, professional competence 
and time which are necessary to the head of a great enterprise. 

On taking office he finds himself immediately confronted with a great number of 
problems, of which he has no knowledge but which urgently demand attention. He 
has no documentation to enlighten him, he seldom has a staff of special consultants 
and he is unable to consult the head of the Government, who is too busy to follow the 
details of current affairs. To add to this he usually finds that he must deal directly 
with some twenty subordinates. Swamped, submerged and unable to make decisions 
from first hand knowledge, the Minister generally abandons the attempt to exercise 
any executive authority and it is difficult to tell what to dread most, his active inter- 
vention or his inaction. Current operations are ensured somehow by directors acting 
independently of each other, like the ministers, without co-ordination. It is a system of 
water-tight compartments. 

Thus it is not a matter for surprise if certain activities which are partly under two 
departments wait long months for decisions which should be made in a week, and if 
those which are under two ministries wait indefinitely. It is also not surprising that 
ministers who might seriously influence the interests of the state, find themselves 
powerless to carry out the least reform. The remedy lies in a better recruitment of gov- 
ernment personnel and in introducing the use of good administrative tools. The Cabi- 
nets also would probably last longer under a better functioning of government, and 
their present harmful instability be lessened. 


The Head of a government activity as in the case of the Manager of a big private 
undertaking, should have the following qualities: administrative ability, professional 
competence and stability. This combination of qualities is sometimes met, but it may 
also happen to be completely lacking. 

On November 4, 1920, I delivered to M. Louis Deschamps, Under-Secretary of 
State, a report which he had asked me to make on the organization and operation of 
the Department of Posts, Telegraphs and Telephones of which he was Director Gen- 
eral. The report began thus: "During the past year I have been studying the govern- 
ment undertaking of P.T.T., and I have noted many faults of administration of which 
the following are the most outstanding: 

1. An unstable and incompetent chief at the head of the undertaking; 

2. No long term plan; 

3. No budget; 

4. Abusive and excessive interference from members of Parliament; 

5. No incitement to enthusiasm and no reward for services rendered; 

6. Absence of responsibility." 


Chance had brought me into contact with one of the great enterprises where the 
usual faults of public administration were most pronounced and I set myself to prove 
that most of these faults are the result of bad management. To remedy these defects the 
Government, citing the Administrative Theory, has recently proposed to set up a com- 
mittee of consultants without changing in any way the method of recruiting the di- 
rectorate. Now let us see, according to my survey of 1920 how this recruitment is 
made: "A ministerial crisis arises. The composition of the new ministry appears in the 
Officiel of January 16. The Under-Secretary of State, Monsieur A., lawyer and 
deputy, holding the office of Director General of the P.T.T. is replaced by the Under- 
Secretary of State, Monsieur B., lawyer and deputy. January 17, Monsieur B. comes 
to the office and meets Monsieur A. These gentlemen talk together for a few minutes, 
then Monsieur A. takes his leave, having already emptied the drawers of his desk. 
There has been no presentation of the heads of departmental activities to the new 
chief and no plan of operations is given to him for his guidance. 

"Thus takes place a handing over of office in one of the most important adminis- 
trations of the Government. The chief who is leaving took office twelve or fifteen 
months earlier without knowing anything of the great undertaking he was called 
upon to direct, and the new chief is in the same position and will probably leave after 
the same interval." A head thus recruited has neither stability nor competence and it 
is highly probable that he is not gifted with administrative ability, nor does he have a 
good staff nor good administrative tools. The Administrative Theory condemns the 
illusion that under these conditions, the administration of the P.T.T. could be im- 
proved by the addition of a committee of consultants. The appointment of such a 
committee is generally an admission of helplessness on the part of the Administration 
and a way of lulling the vigilance of those interested. Even if all the subordinate per- 
sonnel were very efficient the undertaking would not function well with a mediocre 
head. It is like an individual all of whose limbs are sound, but who has a diseased 
brain. We must realize that while all possible reforms may be carried out under a 
good management, no serious reforms can be carried out under a poor management. 
The Administration of the P.T.T. evidently does not understand the Administrative 

This is not the case in another great French Administration — the Army. 
Planning and organization are constant preoccupations with our great military 
chiefs; authority, discipline, unity of management, unity of command, and subordina- 
tion of individual interests to the common good are constantly kept in mind. The Na- 
tional Military School has received the Administrative Theory with favor because it 
recognizes in it a synthesis of the ideas that it advocates. But these ideas have not yet 
reached the ministries, and so the abuse of written communications continues to 
prevail, and a Minister has twenty assistants where the Administrative Theory says 
that a manager at the head of a big undertaking should not have more than five 
or six. 

The persistence of faults of organization in the directorate of the P.T.T., taken as 
an example, gives an idea of the difficulty presented by the reform of the public 
service, and this difficulty has recently given rise in several countries to special 
organisms called Reform Commissions, or Economy Commissions. 

the administrative theory in the state 111 

Councils for Improvements 

The Administrative Theory supposes that in every great enterprise there is a 
permanent council for improvements whose function it is to make researches on all 
possible improvements in the enterprise and to carry them out under the auspices and 
authority of the director. 

An organism of this kind seems to me indispensable in order to study and carry 
out reforms in the Enterprise of Government, which perhaps, more than any other, is 
in constant need of them. 

A firm determination and continued action are needed to overcome the resistance 
that ignorance, routine and individual interests oppose to reforms. Temporary mani- 
festations in which the higher authorities take little part can have no important re- 
sults. Continued action requires a special permanent organism. 

The firm determination of a Head, such as the Prime Minister, must be based 
upon a profound conviction of the need for reform and on an accord with the Presi- 
dent and with Parliament. This accord in turn must have the support of favorable 
public opinion. 

M. Frangois Marsal, Minister of Finance, appointed in 1920 a High Commission 
of Inquiry to look into the question of possible retrenchments in the budget. M. 
Maurice Bloch, Procureur General of the Cour des Comptes was chairman of this 
commission, which was succeeded in August, 1922, by the Economy Commission 
(Commission Superieure des Economies) under the chairmanship of M. Louis Marin. 
The Commission has found many defects of organization and function in most of the 
government services, but when its work is finished in a few months and its findings and 
recommendations are embodied in a report, what will be the fate of the latter? We 
have seen that the Prime Minister is much too busy to study a voluminous and com- 
plicated document requiring a great deal of special knowledge, and we also know 
that he has no group of men around him whose duty it is to perform this task for him. 
The Ministers are too absorbed by their routine duties and too uncertain of their 
tenure to devote themselves to difficult studies which take time and whose recom- 
mendations they cannot put into effect, and which can only create embarrassment for 
them. As for the Directors, whose tranquillity has been disturbed by the Commission 
and who have sometimes been severely taken to task by it, they usually have no sym- 
pathy for the reforms it proposes. 

Under these conditions it is highly probable, not to say certain, that this attempt 
at reform will be futile like most of those which have preceded it. 

It would be otherwise if there were a permanent Council for Improvements or 
Reforms associated with the Prime Minister. This council would have for its mission to 
guide and direct studies on reforms in the various departments and offices and to see 
that they are carried out. The creation of such a body should be carefully studied by 
competent authorities and the following set-up is suggested only as a tentative 

The Council for Improvements should be composed of five members: 

The first would have charge of planning, organization, command, co-ordination 
and control, or in other words, of administration; 


The second of financial problems; 

The third of the organization of accounting and statistics; 

The fourth of legal matters; 

The fifth should be a business man. 

The Council for Improvements should be represented in each government de- 
partment by a liaison officer subordinate to the Minister, and particularly charged 
with the duty of keeping him in touch with the studies being made in his department. 

The Council itself should be in charge of the improvements to be made in the 
machinery of the superior services of the state. 

In the work of governmental reform the directing powers of the state need to be 
helped, sustained and encouraged by public opinion. Public opinion, however, is not 
prepared for this: unenlightened upon the projects for reform and ignorant of the 
benefits to be derived from them, it remains indifferent and sceptical. I believe it 
could be aroused by imbuing it with the idea that there are important reforms both 
necessary and possible, and that among their results might be a considerable reduction 
in taxes and a very appreciable decrease in the cost of living. 

Public opinion may be guided in the direction of reform by a knowledge of the 
principles and rules of the Administrative Theory and we can help to spread this by 
teaching it in the institutions of higher learning. But the effect will be slow and we can 
hasten it by immediately putting into use administrative tools (outillage administratif) in 
all the public services. 

At the same time we can endeavor to arouse public opinion by showing the eco- 
nomic advantages which should result from the projected reforms. It is for this reason 
that I present to the Congress the following estimates which do not pretend to be 
rigorously exact, and which for me are only a means of interesting public opinion. 

I have endeavored to put into figures the pecuniary interest which the Manage- 
ment of Government has for us. This interest is composed of two elements, one of 
which is easily determined — it is the cost of government as represented by its annual 
budget plus its extraordinary requirements (immovable assets, supplies, etc.). If the 
administration is efficient, these expenditures, however high, are justified — the 
country has its money's worth. If on the contrary the administration is bad, the ex- 
penditures constitute a total loss. But we rarely find that either term can be applied 
sweepingly, for we seldom find an administration completely good or completely bad. 
The truth is that good and bad are mixed in varying proportions. Is it rash to assert 
that under the best administrations there is room for reforms that would lower the 
cost of government without curtailing its services? What then shall we say of the other 
administrations? Taking the public services as they are, the good with the bad, in 
order to form an idea, can we not estimate at 10 per cent of the ordinary and extraor- 
dinary budget, that part of the expenses of administration that should interest public 
opinion, in the sense that the country will spend it or not according to whether the 
business of government has been efficiently managed or otherwise? 

But this is only one element of the situation. The second is the repercussion, good 
or bad, that the operation of the government necessarily has upon the general eco- 
nomic life of the Nation. We cannot possibly estimate this in figures, but every one 
knows that it represents a huge sum. Taking, for example, the French government 


service of the P.T.T., where the capital actually invested is estimated at one billion 
francs, the coefficient of 1 per cent would show an economy of 1 00 million francs as the 
advantage of a good management over a bad one; and when the proposed future 
capitalization of two billion francs has been effected, there would be an economy of 
300 millions, without counting the much more important advantages that would ac- 
crue to the country as the result of a better administration of the service. If this calcu- 
lation were extended to all the activities of the state, the result would be a great many 
billions, and billions of economies means reduction of taxes, more abundant produc- 
tion and decreased cost of living. 

The High Commission of Inquiry (Comite Superieur d'Enquete) under M. 
Maurice Bloch, estimated at about two billions the amount of savings that could be 
realized without any important administrative reorgnization, by simply cutting down 
appropriations and demanding more efficiency from the various government services. 
The High Commission for Economy (Commission Superieure des Economies) under 
M. Louis Marin estimated that a complete reorganization of the Government under a 
new plan would result in an economy of four or five billion a year. Such, according to 
this estimate, is the cost to France of a bad management of public affairs. This is not 
in contradiction with the estimate of the Commission of Enquiry nor with that I have 
made above for the P.T.T. If we add to this the inconveniences of every kind that are 
the result of an inefficient government and the money the inhabitants failed to earn 
for the same reason, we are led to double the above sum and we find ourselves con- 
fronted with the formidable figure of eight or ten billions. 

In a country of less than forty million inhabitants, this sum represents a loss of 
more than 200 francs to each person — 100 francs lost and 100 francs not earned. To 
the average family consisting of parents and two children, an inefficient government 
costs more than 800 francs a year: the half of this sum that corresponds to the actual 
loss, should be deducted from the family income, for it goes to the state for taxes. The 
other half affects the very basis of the family fortune, for it applies to wealth in process 
of formation, whose development is paralyzed for individuals, without any immediate 
or remote profit to the state. It is true that this estimate cannot pretend to be really or 
even approximately exact, for the income and wealth of the inhabitants are not uni- 
formly affected by inefficient government, but if the mean indicated is necessarily not 
exact, the calculation that leads to it is based none the less on the uncontrovertible 
principle that all French citizens bear a part of the expense of government and have an 
interest in the good management of public affairs. 

Whatever their economic status all French citizens bear a part of the expense of 
government. Even though taxes on wealth and income reach only a fraction of the 
taxpayers, taxes on commodities strike all without distinction. All of these taxes could 
be reduced if the needs of the state were less, and especially if there were effected that 
indispensable condition to any important economy, the reorganization of the High 
Command of the Public Service. We are therefore forced to maintain that the mass of 
citizens, and not only the most fortunate, bear under the guise of a reduction of in- 
come the disastrous effects of our inefficient administration. 

The taxpayers of all countries have an interest in the good management of public 
affairs, and hence in administrative reform. The abusive levies effected by the state 


directly or indirectly on the revenues of capital and labor form an obstacle to thrift, 
hamper the formation of new wealth and in consequence rob production of its means 
of development. Industry, agriculture and commerce are not alone to suffer from this 
state of affairs: the consumer bears also the unpleasant consequences, since every re- 
striction of economic activity is manifested by diminished comforts and a rise of prices. 

These considerations seem to me of a nature to interest the public in administra- 
tive reforms and to encourage the Managers of the State to study them and put them 
into effect. I am anxious to have the Congress share this opinion. 

What is my conclusion? It is that the essential condition for a successful operation 
of the Public Service is a good High Command, and a good High Command entails a 
good staff and good Administrative Tools. 


With Special Reference to the Work of Henri Fayol 


L. URWICK, O.B.E., M.C., M.A. 

Chairman, Urwick, Orr and Partners, Ltd. 

Consulting Industrial Engineers 


Director of the International Management Institute 





With Special Reference to the Work of Henri Fayol 

This paper has a threefold purpose: 

I. It calls attention to the work of a famous French industrialist which is per- 
haps too little known in this country. More than any other European who has lived 
in this century, Henri Fayol is responsible for directing minds to the need for studying 
administration scientifically. He has laid down broad lines which no subsequent 
student should neglect. 

II. But his logical analysis of the operations involved in business and partic- 
ularly of the function of Administration stops quite suddenly when he begins to talk 
of principles: he becomes empirical in his presentation and attitude. This sudden 
check in his thought will be shown to be due to the "practical man fallacy," inevi- 
table in the light of his background and experience. It is an intensely interesting 
example of the limitations imposed on scientific study by immediate administrative 

III. Despite his own practical attitude and refusal to consider a logical arrange- 
ment of the principles, it will be found, on closer examination, that his general treat- 
ment of administration can be presented as a complete, logical scheme. Within that 
scheme are two subsidiary systems. The one deals with the structural and the other 
with the human aspects of administration. The former has been arranged to corre- 
spond with the scheme of the principles of organization developed by Mr. Mooney, 
President of the General Motors Export Company. The latter has resulted from ar- 
ranging the balance of Fayol's principles and administrative duties on the same gen- 
eral lines. 

Three points of interest arise from this study: 

(A) Despite the fact that Fayol's work was, apparently, not known to Mooney, a 
statement of principle corresponding to each head of Mooney's analysis is discovered 
in Fayol's empirical lists. These principles deal with structure. 

(B) The remaining principles do exactly correspond to a second scheme dealing 
exclusively with the operating or personal aspects of administration. That constitutes 
fairly effective proof that Fayol's refusal to consider structure and operation sepa- 
rately was, in fact, the cause of his failure to push his very logical handling of the func- 
tion of administration to its conclusion. 

(C) Despite this avowed refusal to continue along logical lines, the general ar- 
rangement presented does, in fact, use up the whole of his principles and adminis- 
trative duties exactly. One, "impose penalties for mistakes and blunders," is used in 
two places. Otherwise there is neither repetition nor omission. The successful, practical 
organizer was being far more logical than he himself realized or was prepared to 

These considerations, if accepted, appear to the author to strengthen greatly the 
general contention that a scientific analysis of the arts of administration and organiza- 



tion can be developed. It is hardly necessary to emphasize the importance of this con- 
clusion, if it is valid, not only for business administration, but for political science. 
Many exact scientists have emphasized in recent years the lack of balance between our 
knowledge of research devoted to the material aspects of nature and our knowledge of 
social organization. The possibility of a science of management, or, as Fayol would 
say, of government, is the only concrete hope before humanity of an ordered and 
satisfactory solution of the economic and social problems which that lack of balance 
has created. 

Henri Fayol was born in 1841 and died in 1925. He qualified in 1860 as a mining 
engineer and entered the great coal and metallurgical concern of Commentry-Four- 
chambault. In 1866 he was appointed Manager of the Commentry collieries, to which 
were added in 1872 the collieries of Montircq. In 1888 he was appointed General 
Manager of the Company Commentry Fourchambault & Decazeville, which at the 
time was in a most critical condition. He was brilliantly successful. When in 1918, 
after thirty years of practical command, he handed over to his successor, the concern 
was on a firm basis of prosperity, endowed with remarkable financial stability and as- 
sets of unusual value. 

His career, thus passed entirely with one organization, is typically European. It 
is in marked contrast in this respect to F. W. Taylor's varied experience. But in the 
scientific enthusiasm with which they approached practical problems, both men were 
remarkably alike. Taylor experimented with high speed steel and the technique of 
cutting metals: Fayol with shafting, timbering, subsidence, and a geological survey of 
the coal resources of the Commentry field. M. C. E. Guillaume won the Nobel 
prize in 1921 for researches undertaken at the Imphy factory with Fayol's assistance 
and encouragement. Both men were scientists before they were managers. Both men 
gave the devotion of their later lives to putting science into management. Taylor 
worked from the individual worker at his bench upwards. Fayol worked from the 
Managing Director downwards. Their contributions are essentially complementary. 

Fayol always insisted that his success was due, not to any personal qualities, but 
to the methods he employed. His later years were dominated by the desire to prove 
that, with scientific forecasting and proper methods of management, satisfactory re- 
sults are inevitable. Shortly before his retirement he expounded his theory of adminis- 
tration in two lectures before the Society for the Encouragement of National Indus- 
try. These lectures formed the foundation of his famous book, "Industrial and General 
Administration." 1 He applied his ideas to public administration in a report on the 
French postal and telegraph services, published under the title "The State Cannot 
Administer." 2 They were further crystallized by the foundation of an organization in 
Paris for their development and teaching — The Centre of Administrative Studies. 

Fayol's thesis, which is illustrated graphically in Table I, starts with the proposi- 
tion that all the operations which occur in business undertakings can be divided into 
six groups: 

1 Except where otherwise stated, all references are to Administration Industrielle et Generate by Henri Fayol, Dunod, 
Paris, 1925. English and German translations were published by the International Management Institute, Geneva, in 
1925. The English translation by J. A. Couborough is entitled Industrial and General Administration. References show the 
page in the French and English editions preceded by a capital F and E in each case. 

2 LTncapacite industrielle de l'Etat: les P.T.T. — Dunod, Paris, 1921. 



1. Technical Operations (production, manufacture, etc.). 

2. Commercial Operations (purchases, sales and exchanges). 

3. Financial Operations (finding and controlling capital). 

4. Security Operations (protection of goods and persons). 

5. Accounting Operations (stocktaking, balance-sheet, costing, statistics, etc.). 

6. Administrative Operations (planning, organization, command, co-ordination, control). 

He defines administration as "to plan, organize, command, co-ordinate and con- 
trol." In explanation of these terms he writes: 

" To plan means to study the future and arrange the plan of operations. 

"To organize means to build up the material and human organization of the business, organ- 
izing both men and materials. 

" To command means to make the staff do their work. 

"To co-ordinate means to unite and correlate all activities. 

"To control means to see that everything is done in accordance with the rules which have 
been laid down and the instructions which have been given." 




("To govern an undertaking is to conduct it towards its objective by trying to make the best possible use 
of the resources at its disposal; it is, in fact to ensure the smooth working of the six essential functions") 
















Finding and 

Protection of 



Sales and 


goods and 

balance sheets, 






making best 




possible use 


statistics, etc. 

of available 


funds and 

avoidance of 






Organize Co- 





To fulfil 


give things 

"After an 

"Seeing that ev- 

| Forecast 

Plan | the 16 

erything is being 


"To fore- "Administra- their 


has been 

carried out in 

is fore- 

see as the tive Duties," proportion 

formed it 

accordance with 


word is v - Table III and 

to adapt 



the plan which 

used here 

the means to 

made to work 

has been adopted, 

means both 

the end" 

and this is 

the orders which 

to foretell 

the function 

have been given, 

the future 

of command" 

and the principles 

and to pre- 

which have been 

pare for it" 

laid down" 

(From "Industrial and General Administration," 

English edition.) 


He adds: "Administration regarded in this way must not be confused with gov- 
ernment. To govern is to conduct an undertaking towards its objective by seeking to 
make the best possible use of all the resources at its disposal; it is, in fact, to ensure the 
smooth working of the six essential functions. Administration is only one of these func- 
tions, but the managers of big concerns spend so much of their time on it, that their 
jobs seem to consist solely of administration." 3 

Thus Fayol begins his theory with a classification along functional lines of the 
main groups of activities to be found in business. He then deals with Administration, 
breaking it down into five main aspects. It may be noted, however, that his term pre- 
vqyance, which has been translated by "planning" in the English edition, really covers 
two or more separate and distinct processes. He says prevoir (literally "to foresee") as 
used here means both to foretell the future and to prepare for it; it includes the idea of 
action. La prevoyance is found in innumerable forms and at many different points; it 
becomes most concrete and obvious in the plan of operations {programme abaction) 
which is at the same time its most effective instrument. In illustrating his idea he de- 
scribes the method of drawing up the plan of operations in a big mining and metal- 
lurgical firm. The details which he gives of these forecasts, particularly on the finan- 
cial side, are in effect a complete summary of the principles which have subsequently 
attained so much prominence in business practice under the title of Budgetary Con- 

His general introduction to the study~of administration concludes with what, in 
the author's view, is one of the most important features of his invaluable contribution: 
"Whatever the function being considered, the chief characteristic of the lower em- 
ployees is the special ability appertaining to the function (technical, commercial, fi- 
nancial, etc.) and the chief characteristic of the higher employees is administrative 
ability. Technical ability is the most important quality at the bottom of the industrial 
ladder and administrative ability at the top." As Fayol observes, "the technical func- 
tion has long been given the degree of importance which is its due, and of which we 
must not deprive it, but the technical function by itself cannot ensure the successful 
running of a business; it needs the help of other essential functions and particularly of 
that of administration." 

Many of the failures to conduct large combinations of firms successfully may be 
traced to lack of administrative knowledge in men who have been highly successful in 
charge of smaller concerns, that is in positions demanding a relatively larger propor- 
tion of technical capacity. A man with high administrative qualifications who has 
never been inside a steel or biscuit works may succeed. A man with all the knowledge of 
steel or biscuit making in the world, but who lacks the requisite level of administrative 
knowledge and capacity will certainly jail. 

Unless and until Fayol's thesis on this point wins universal acceptance, avoidable 
losses will occur. They will be due in large measure to failure to recognize the ap- 
propriate balance of qualities in those selected to fill the higher executive positions in 
business. At the same time the arts of administration and management will receive 
less study than they deserve or than is necessary for the training of future generations. 

F. p. 59; E. p. 35. 


So long as technical capacity and experience are treated as superior qualifications for 
higher posts, it is inevitable that administrative ability should be regarded as an 
empirical and secondary consideration, something the technical man "picks up" in 
the course of his career. 4 

At this point, however, Fayol abandons any further attempt to analyze adminis- 
tration logically. Of the principles of administration he writes: "There is no limit to 
the number of principles of administration. Every administrative rule or device, which 
strengthens the human part of an organization or facilitates its working, takes its place 
among the principles, for so long as experience proves it to be worthy of this important 
position." He then lists fourteen of such principles, summarized in the second column 
of Table IV, and adds: "I shall leave the review of principles at this point, not because 
the list is exhausted — it has no precise limit — but because it seems to me particu- 
larly useful at the moment to endow the theory of administration with about a dozen 
well-established principles, on which public discussion can conveniently be focussed." 
The reason he gives for abandoning a logical arrangement is a sound one. He wishes to 
insist on elasticity in the application of administrative principles: "The same principle 
is hardly ever applied twice in exactly the same way." This is a fundamental concept 
in administration. His discussion of the principles is extremely suggestive and illumi- 
nating. But, it is open to question whether the reason which he gives, sound though it 
is, was his real reason for leaving his logical analysis aside. 

When he comes to deal with organization as an aspect of administration, he 
does not distinguish clearly between designing the structure of an enterprise and pro- 
viding the personnel to carry out the various groups of activities thus demarcated. 
The phrases which he uses, organisme social and corps social, are virtually untranslatable 
in English. They have been rendered by "social organism" and "the staff as a body." 
The term organism is, however, always used of a living entity. And it is clear that he 
regards organization wholly from this standpoint. 

He says, for instance, that the staff of an enterprise as a body is frequently com- 
pared to a machine, a plant or an animal. He objects to the first comparison on the 
ground that it might leave the impression that "like the parts of a machine, the mecha- 
nism of administration cannot transmit power without some loss. This is not correct; 
the mechanism of administration, that is, every intermediate link in the chain, can and 
must be a source of energy and ideas; there is in each of these links, or men, a power of 
initiative which, if properly used, can considerably extend a manager's range of ac- 
tivity." 6 On the other hand, he is in considerable sympathy with the two organic com- 
parisons and particularly the second — "the part played by man in an organization is 
analogous to that of the cell in the animal." 

Indeed he writes somewhat slightingly of any attempt to isolate the structural 
factor: "If we could eliminate the human factor, it would be easy enough to build up 

4 Fayol's insistence on the importance of administration should not be taken to imply any justification for the cre- 
ation in any enterprise of a special administrative group or class. A situation such as that revealed by the Bridgman 
Report on Post Office Organization in Great Britain ( 5 ) where the chiefs of the technical functions occupied a secondary 
position, and were separated from the chiefs of the operating divisions and from the administrative chiefs by a special 
Secretariat or Headquarters, inevitably leads to conservatism and lack of enterprise. Decisions tend to become merely 
administrative. Responsible executives are insulated from the discipline and stimulation of direct contact with technical 
development. Cross-functionalization breaks down. 

s C.M.D. No. 4149. Committee of Enquiry on the Post Office, 1932 Report. 6 F. p. 83; E. p. 46. 


an organization; anyone could do it if they had some idea of current practice and had 
the necessary capital. But we cannot build up an effective organization simply by 
dividing men into groups and giving them functions; we must know how to adapt the 
organization to the requirements of the case, and how to find the necessary men and 
put each one in the place where he can be of most service; we need, in fact, many sub- 
stantial qualities." 7 

Just because Fayol was a practical man of long experience, this attitude was 
probably inevitable. The responsible administrator cannot divorce himself from the 
human factor. Enterprises are built up of living persons engaged in a joint undertaking 
and thus partake of the character of living organisms. To regard them as purely 
mechanistic is to open the door to every kind of error. But in so far as it is claimed that 
Fayol' s work establishes a theory of administration — a claim which has certainly 
been preferred by his disciples — his concept of organization sets limits to his 

It is impossible for humanity to advance its knowledge of organization, unless the 
factor of structure is isolated from other considerations, however artificial such an 
isolation may appear. Existing confusion in the world's affairs makes such an advance 
in knowledge essential if social order is to be preserved. It is just because responsible 
administrators of all kinds, who inevitably carry the greatest weight with public opin- 
ion in such matters, cannot, owing to the nature of their task, achieve such an intellectual detach- 
ment, that our progress in the sociological sciences lags behind our growing command 
of physical resources. 

The statesman cannot consider what is the most effective form of organization as 
a technical question, without at the same time taking into account the traditions of 
local patriotism, the activities of the press, the current emotional tendencies of the 
population, all the thousand and one contemporary influences which may constrict 
what is desirable to what is possible. The captain of industry administering a vast 
enterprise should never lose sight of the quality and personal idiosyncrasies of the 
managers at his disposal, of the temper of his workers, of the drift of public demand for 
his own and allied products. Because they are "practical men," because it is their duty 
to make the best they can of a complex human situation, they would be doing less 
than their duty if such considerations did not form a constant background to their 

Yet, it is only by achieving the isolation of one or more factors in a problem that 
science advances its frontiers. Medicine is concerned with the preservation of life. But 
it gains knowledge from anatomy. The art of administration demands knowledge of 
the human factor. But that does not mean that it is impossible to study the structure of 
enterprises anatomically, to isolate organization and treat it as a technical problem. 

Fayol's analysis of organization starts with a list of the administrative duties 
which every organization shall be designed to accomplish. This list is shown in the 
first column of Table IV. It does not correspond exactly with his list of principles. But 
the major discrepancies, where there is no principle corresponding to an administra- 
tive duty, are covered by the fact that there is one of his aspects of administration 
corresponding to each such administrative duty. 

7 F. p. 83; E. p. 46. 



Having noted these characteristics of Fayol's exposition, let us turn for a moment 
to two authors, who have made an attempt to analyze organization considered as 
structure. In 1931, Messrs. J. C. Mooney and A. P. Reiley published, under the some- 
what misleading title "Onward Industry," an analysis and historical comparison of 
the principles of organization as manifested in governmental, religious, military and 
industrial institutions. Their general scheme of principles is outlined in Table II. The 
logical standard which they set themselves was a high one. It was founded on Louis F. 
Anderson's Das Logische, seine Gesetze und Kategorien. Felix Meiner, Leipzig, 1929. 
Broadly speaking, it postulates that every principle has its process and effect. But, if 
the principle, process and effect have been correctly identified, the process and effect 




1. Principle 

2. Process 

3. Effect 

1. The 


Authority; or 

per se 



2. The 




3. The 








(From "Onward Industry") 

will each be found to have their own principle, process and effect. Thus the whole 
statement correlates into a complete square of nine items. 

Without going into detail, their table may be expanded as follows. The purpose of 
all organization is to unify effort, that is co-ordination. Co-ordination finds its princi- 
ple in authority. Authority enters into process with the scalar chain, that is the hier- 
archy or "line," as it is sometimes called. Its effect is the assignment and integration of 
functions. The scalar process has its own principle in leadership. But leadership can 
only enter into process with delegation and its final effect is always the definition of 
function. Somewhere at the end of all delegation, what is delegated is responsibility 
for the performance of a particular task. In all forms of organization, three main func- 
tions can always be distinguished — the determinative or legislative, the applicative 
or executive and the interpretative or judicial. These are related as principle, process 
and effect. 

In comparing Fayol's work with the theory of Mooney and Reiley, it occurred to 
the author that if the apparent empiricism in Fayol's presentation is, in fact, due to 
the "practical man" fallacy, to a refusal to distinguish between structure and opera- 
tion, then the introduction of this distinction as a basis of classification might reveal a 


much greater degree of logical coherence in his principles and administrative duties 
and a much closer correspondence with the more formal scheme of Mooney and 
Reiley than appear on the surface. In short, the claim that he has laid down the out- 
line of a theory of administration would be largely substantiated. The author believes 
this to be the case. Fayol's immense practical experience, while it tempted him to 
ignore the importance of logical analysis, secured him against dealing with inessentials. 
And as a consequence of the unity of principle which is fundamental in all effective 
administration, his statement of essentials, while unco-ordinated in form, does in fact 
lend itself to a completely correlated arrangement. 

In the first place, his general analysis of the function of administration divides it 
into the five elements of prevoyance, organization, co-ordination, command and con- 
trol. But, as has been pointed out, the term prevoyance really covers two separate con- 
ceptions, to foretell the future (forecasting) and to make a plan to meet it (planning) . 
It will be noted that these six aspects of administration readily fall into three groups of 
two, related to each other as process and effect. Thus the process of forecasting results 
in a plan: organization has for its purpose and effect co-ordination: command issues 
in control. At this stage of his argument Fayol was not concerned with underlying 
principles — only with the processes of administration. If, however, it is possible to 
discover the underlying principles in other parts of his work, it is legitimate to apply 
them to complete the scheme. 

The underlying principle of any form of administration which aims at scientific 
precision and integrity must be investigation or research yielding information. Ad- 
ministration must proceed on a basis of facts. While Fayol nowhere states this cate- 
gorically, it is implicit in his discussion of planning. He defines the characteristics of a 
good plan of operations as unity, continuity, flexibility, and precision. Of this last qual- 
ity he writes: "A plan must be as precise as is possible in view of the unknown forces 
which influence the results of an undertaking. As a rule, it is possible to trace the line 
of action for the near future with quite a considerable degree of accuracy, while just a 
general guide is enough for operations which are some distance ahead; before the 
time comes when they will have to be carried out, more information will be available, 
which will allow the line of action to be more accurately determined." 

The process of forecasting is reflected in process by the establishment of an organi- 
zation. And Fayol clearly indicates the principle which should underlie organization 
in his second administrative duty: "See that the human and material organization 
are suitable for the objects, resources and needs of the undertaking." This principle 
may be termed appropriateness. It is a fundamental consideration which must qualify 
the purely intellectual process of forecasting before it takes shape in human and 
material arrangements for action. 

Similarly, a plan, however prudent, accurate and farsighted, remains a plan, 
a project, until staff are set to work to carry it out. That is to say, it enters into process 
with command. But it is possible to conceive a plan without action, just as it is 
possible to conceive of a great quantity of commanding without a plan. The reason 
for a plan, the conception which links it with command has also been stated by 
Fayol in his tenth Principle and fourteenth Administrative Duty — "ensure material 
and human order." 



But if these three principles — investigation, appropriateness and order — are 
added to the six elements of administration, there is immediately the outline of a 
logical scheme for the function of administration as a whole, similar to that developed 
by Mooney and Reiley in relation to the principles of organization only. Thus the 
fundamental principle of administration, which is investigation, enters into process 
with forecasting which takes effect in a plan. Forecasting has its own principle, 



Forecasting, Planning are the two parts of Fayol's Prevoyance. 

Organization, Co-ordination, Command, Control, are the other four original elements. 

(Investigation), (Appropriateness), (Order), are principles found or implicit in other parts of 

Fayol's exposition. 

1. Principle 

2. Process 

3. Effect 

The function of 

1. Administration 




must depend on 

which is implicit in 

which is part of 

which is the other 

the basic principle 

Fayol's discussion of 

Fayol's "Prevoyance," 

part of Fayol's 

of Investigation 

Planning, and enters 

and results in 


into process with 



2. Forecasting 




has its own 

"See that the human 

which has the effect 

Cf. "The internal 

principle in 

and material organ- 

of securing 

objectives of 

Fayol's 2nd Ad- 

ization are suitable." 


Organization are 

ministrative Duty 

It enters into process 

co-ordinative always" 


with Organization 

(Mooney and Reiley) 

3. Planning 




finds its principle 

"Ensure material and 

and its effect is 

in Fayol's 10th 

human order." It 


Principle and 14th 

enters into process 


with Command 

Duty, viz.: (Order) 

process and effect in appropriateness, organization and co-ordination. Planning finds 
its principle in order, enters into process with command and results in control. 
This scheme is shown in diagram in Table II. 

If, however, Table III is examined closely, it will be noted that read horizon- 
tally its second and third columns correspond closely with the distinction which has 
been drawn between the structural and operating aspects of administration. Fore- 
casting, organization and co-ordination are abstract processes. They consist in 
determining what is wanted to achieve the objective of the enterprise in the given 
circumstances, what groups of activities are required to provide what is wanted, and 
how those groups should be related. They ignore the human factor. Planning, com- 



Administrative Duties 


(F. p. 78; E. p. 42) x 

(F. p. 28; E. p. 19) x 


See that the plan of operations is carefully pre- 


pared and strictly carried out 

(Fayol's "Prevoyance") 


See that the human and material organization 


are suitable for the objects, resources and needs of 

(Fayol's "to organize") 

the undertaking 


Establish a management which is competent and 


has singleness of purpose 

(Fayol's "to command") 


Co-ordinate operations and efforts 

5. Unity of management 
(Fayol's "to co-ordinate") 


Make decisions which are clear, distinct and 



Make careful selection of staff — each depart- 
ment has a competent and energetic head: each 
employee where he can be of most service 



Define duties clearly 



Encourage the desire for initiative and responsi- 

13. Initiative 


Reward men fairly and judiciously for their 

7. Remuneration 


Impose penalties for mistakes and blunders 

2. Authority 


See that discipline is maintained 

3. Discipline 


See that individual interests do not interfere with 

6. Subordination of individual interests 

the general interest 

to the common good 


Pay special attention to unity of command 

4. Unity of command 


Ensure material and human order 

10. Order 


Subject everything to control 

8. Centralization 

9. The hierarchy 
(Fayol's "to control") 


Avoid red tape 

14. Esprit de Corps 


References to Fayol's "Administration Industrielle 

Principles listed by Fayol for which he 

et Generale": French edition shown' 'F"; English 

does not show equivalent administrative 

edition shown "E" 


1. Division of labor 

11. Equity 

12. Stability of staff 

mand and control, on the other hand, are concrete. They are processes of realization 
or operation. They are intimately concerned with obtaining the best results from 
the human factor. 

If this point is conceded, three consequences follow: 

(A) A complete statement of the principles of administration calls for the addi- 
tion to the table of two subsidiary logical schemes, the one giving the principles of 


organization and co-ordination and concerned with structure, and the other the prin- 
ciples of command and control and concerned with the human factor. 

(B) If Fayol's empirical lists of principles and administrative duties are really logi- 
cal and complete, it will be found that one of his statements corresponds to each prin- 
ciple, process and effect in these subsidiary schemes and no more. There will be no 
principles or administrative duties superfluous and every aspect of both schemes will 
be provided for. 

(C) If Fayol's thought is consistent with the principles of Mooney and Reiley 
it will be sufficient to adopt their scheme to cover the structural aspect of adminis- 
tration and to show where Fayol has corresponding principles or administrative 

This correlation has been attempted in Table VI which is an expansion of Table 
III to include the two subsidiary systems of principles. I may mention that there is 
no Table V. 

Following Mooney and Reiley, organization and co-ordination are treated as 
the processive and effective aspects of the same structural activity. Command and 
control are similarly treated as the processive and effective aspects of the same op- 
erating activity. In the first case Mooney and Reiley's scheme of principles has merely 
been inserted, and the appropriate statement from Fayol's lists placed under each 
heading. In the second case a fresh scheme of principles based on those of Fayol's 
principles or administrative duties which bear upon the human factor has been 
constructed. In addition, where any of Fayol's principles or administrative duties 
were merely restatements of his original six elements they have been put in the ap- 
propriate place in the main scheme. 

The principles of command and control are those which bear upon the human 
factor. The underlying purpose of command and control is to secure that all activi- 
ties subserve the common interest. The principle expressing this purpose is centraliza- 
tion. It enters into process with appropriate staffing and its effect is esprit de corps. 
Appropriate staffing depends on the principle of selection and placement. It enters 
into process with the stimulation of the individuals selected by rewards and sanc- 
tions. Its effect is to release each individual's initiative in the interests of the under- 
taking. Esprit de corps depends primarily on equity. It enters into process with the 
enforcement of discipline. Its effect is stability of staff. 

On the whole the table substantiates the view of Fayol's Principles and Admin- 
istrative Duties suggested above. The scheme as a whole is logical and self-consistent. 
All Fayol's statements are included. There are no gaps and only one Administrative 
Duty (No. 10) is used twice. Administrative Duties 1 and 8 are each divided into their 
structural and operating aspects. There are no superfluous statements. The subsidiary 
schemes correlate with the general scheme, and, with one exception, with each other. 

The exception is interpretative functionalism. The fact that it is omitted, that 
Fayol has no operative principle or administrative duty corresponding to the judicial 
function in government, strengthens the argument as to the "practical man" fallacy. 
It is exactly the discrepancy one would expect. As Mooney and Reiley have pointed 
out, there is a general failure to identify and isolate the judicial or interpretative 
function in business organization. Even today the business executive, like the primi- 

— 8 

a a> 



h a 

H I. 

& O 

■ a 

— O 

a *> 




»b e ai 

e K ►> 
Hrt a — 
vati t*** 

a bh e to 

»< Q.J5 «••* 

a a a s h 

• • 041 O 

•^— a 

BO*. 4i -I 0. 
*• Ko-Jd B 

* S *»« K 
Jrfl.v. -S.O 
SO ^rf-rj 

ft.** 8 • 
jo a otj a*o 

HH • 4» a 
C 9»*» OH o 

•Hfl«t9 P.® 

•* o o a 

n • a 

jo — t? 

< ** 4, ).« 

ft" a. e- • 

MttJ* M 

a- Ot. • o 

■* o • • 

U9U H N 

•o o o * 

a*: o — fr» 

u h 0] a e 

o*> i. ("H 

-j a 3 

m — tj «- a 

«- b et 

« o a a 

d h ** *» 

ft* h O O 

** a. 9 9 

a o*(r 

-4 B 

a • hi* 

-4 KB B B 

B BO 3 J3 

.9 H-»*» B 

*r h** o a 

o ••< a 

ft- (CM 

o — ■ 

•e s 

B B B 

-c hHI 

B o I* B 

►> BO B 

a oi-o 

«» 1J 

■ a> 

B — > *» 


H ft* • 

C ► 
K *■* O 

a ~, — 

ft) C 

1 TJ x 




• o 




*» — 


v a 





X a** 
« ** o 

-j a 



u a** 

a e 
^ oo 
x ho 
o v u 

at o 
■>> a 
— a 
u OJ0J 

m**h o 

O « 9 
"<-< 4* 
OiJ B 9 

SO B-*. 

O > o o 
i*. a bjg 

«« x a 
BAB t. o 

— o — o,— 
a:*.** 4* 

HI ISO B «• 

H a— u - 

"» ►> •»•• B 

O toll. -4 

m s«t« a 

e-t sxo — 

0J4,53**,9 B 
M O»0>4* x 

e» ob < 
m a Of a 

HJI.Pt «J4 
•» BO*i o o 

a • <- 
Jo oo It o 
H4» ft»H 
a**** OOP. 

e-ri (tfl ^ 
IV.** o 

I* t* 44 

44 44W B 

OB— O 

ow« a 

a*»s> o 

OX o 9-4 

*»■ ox** 

.3 -OH 8 

a*« ao t* 

as— o 9** 
O jo o o a 
hbo — 


o *ua**ja — 
■ ► » •> ■ 
do a « 

ft, to*. — J«» 


He no v 
>•* o o ► t. 
HI ■ ohh o 
Hft* o Jd **•*. 

CO**** ojo 

a c toil 
m t* 9« o a 
tm a**- M o 
m> sax 
sow •* — 

SMS BM 10 
— a *i 

** ♦* 6i 

3 00PO 
BS C« h 
H S «4*B 

ill eh 

■ X « B 

O— O OJ3 
-»tJH w 
•» • s 


hd o— t. 
**♦* o ■ 

m a 
— 4. n*o ts 

tl*>— I. » 

1 B <• 

■ atlri 

■O—i o o o 
« t —*• a. 


O ».- 
CC fi*l 
O— o 

o jo a 






SB 9 
0* o 



« ». 8 
o CO H 
•H O — 
H o— o 

^*»B B 


•= OO 

•< Sh 

a i.o.« 




a ao 

Rlf*> **t 
faO ** 

« 9 9 JO o 


en o- a — 

W B4*OV N 

gbjo caja a o 

W H*> O < 
H 41»« B — 
<XBt* • 

M*# HO*] BO 

a tow o 
0,4**4 ax 


O.«*r4j0 I. ** 
O. (8 B 9 I* 

«* o— ** o o 

— O I.-* B « 
DQ*> 9 O 9 
t Ith? 

OH o, 

X *. o 

*»o h 

■o B> 

** 9 a* 

m o a 


































9 r- 

















A»— ■ 




a o 




"3 ft* 


•> e 




O o 


















rJ- -* 

*. B 



a u 
o o 
B B 
t I 

B— 1 

a h~ 

H 4*H 


a a 
•u a o 

. s o 

J3 — -o 
*» *» B 

« t* It 
O O b 
X O o 





— o 

o— ~ 

— **ir> 
OH 4 

c h 

a o 


Cq o 

O M 



k£ > 

Ort 4 *H 
8° » 

£« a 

9 O 

4 n o 


a a «9 

II 4 
O t)**v 

o wjat 


jo q**e 
_ B)9«6 

go BOiB* 

s •■** 


**j»jq k 

B B o** 


BW »3 


» St. 

° St. 



B — 

H O 


B Q. 
T3 — 

9 O 
H 9 

U K 


ill. k 

«* B O 

■o U 

** a *. 

H 9 o 

a ■ 


"• Uo 
-» •*» 

a o-o 
ft. o** 

»- ft* 

C ft. D. 
Cj <h ft* 

- ft*. Cb 

a m a 
■-*-* «o 
oi hjb a 

CO a) Mux 

ty ft* » v- 

u *> 0»- 

o e ft. « 
> •**• 
ft. w** o 

O«*(o * 
Mftt t 

C • © • i 
C&WJ3C ■ 

■-»«« -* e 
uuhot. r 

Qffl«»* * 
MbV^W M . 

O a A0' 
k i-l-W I- , 
faO DO** 

tr-H>*a a 

Qj Of* - 

tj« t3ft«C) 

tO 01 

e eft*: 
a •-< © 
o.a e«>. 

I *» 

-v a» © 
(J c i*o 

2 *•• 
lH ».*» 

iv 3 K 

* 3**-4 


K •*+*(- 

J-o Off* 

if o c ato 

c- O M" 

o 5 .»•£ 

; « oj3 
« ft*«> 

CO *-• 

s i a. 



—) 4, - 
» fl*JO 
3--« Q 

S • 

** 04)1* 


a o** 
a o*-> 
«, aa 
ft* w*« 

■ o 

I • 01-1 

-3 ft>*Q 


©*■»© w 

«: 0i 

0*8 00 
•«• «*O0 

0*-t 000 
ftt ft.JLlH<W 


frj «-* 
*- ft*>> 

«e o». 

C**« I. 
» • 9 O 

t3 >-*-.*- 
ft. H-4 

a « «— 

gft* 3* O 
a o a*** 


4. C 4>*J 

c o so a 


a, o a *» to 

O BV. 3 

CJi-t O O 

-, a> ft* 

OA «V 

SJr-t — 

•3 «C« 

c*- ao 

O O I 0*4 

vj b, — • O 

w 0*- 0C O 

O VV<£ ft« 

O « O P. 

K-, •** 4.-. 

a, M a ft* i» o 

en ft«*» 

X w.c a> c 

O * C-H 

Cl O C ■" 

C 0— ' O 

•-* i«a a 

*S 0>tJ 

a ♦* ft. c o 
*- a * * c 
CO c a-H 
e 0o a 
a.** •>• 

«- hMJ «0 
w 0.*>to C 
ft* CO o 

Q,^J 0->1*«« 

O **— *W 

>-«ei o 
o, o tvo 
0,0 a 0H 

^ (U O0 

> oja 

= .• 

en 5" 

n tri 
»- J3 



I •» 
B* I <c 

oh a 


H ► - 

a at 

« c t a 

C JOB) O. 

^* BOH B 
*. 9B*t 

*• U •• « 

U J3H • 

e> 4*H t* 

v* o 
ca a«i «*• 
a- o 


*. B t. 

b-h a 
o t I 

3 « 

«0 9 

H O 

« - X 
p.« H 
*- B H 
O p. t. 
Cj t. 8 

O B 
4» O H 

*, *l V 

o) h a 

fa] 9. O, 



•» »«rl 

K O0 

** B *M 

_OH nt 
O H at* 

BOO -■- 

m a ***B> 
h it a 


<•» BM 

SB a 
B *• 

Sh o t. 
o r> « 

• jo *» i 

N B S 

** B 

« -•» 

a o -** 


*• < 


*»■** t* 




KO k* o 



O M 94* 

Bh >, 
»*8*> B 
JO KB a 

*• 9 9 O 
9 OH (H 
* *"0 O O. 
OtStt H 
J3) o O ft* O 

BO Jfl ft. 
■«*l*i O. 

9 8 

b a 


nt tt 

S ft*-* 


w J 

e o 
•*> o 


jj** * 

3 4* **-■ 
3 ft, ft* 

< ft* 

* O Bid 





H >* 


a is d 

o OO 

ft* o 

■0BSJ1 «rt 

o o 
a +**- 
a ft* 
B a 

O r4W 
O « O 


tive king, is law-maker, administrator, police, judge and jury all rolled into one. 
The extended use of industrial courts is a sign of reaction against this position. It 
is also interesting to note that in one factory where a joint appeal committee was 
established to deal with disciplinary cases, the members chose as their neutral chair- 
man, the company solicitor. A number of dismissals were quashed by the committee 
on the ground that the evidence that the defendant had in fact committed the mis- 
demeanor alleged by his manager was insufficient to satisfy a court of law. 

Because Fayol was relying on his "practical experience" as an industrialist, it 
did not occur to him to question this absence of any separate judicial function. 
It was universal in industrial practice and only becomes anomalous and manifest 
when industrial organization is compared with military or governmental examples. 
In this connection, however, it may also be noted that equity is given as the basic 
principle of esprit de corps. In so far as the workers feel that this identity of executive 
and judicial functions in one person is inequitable, to that extent will business ad- 
ministration fail to achieve complete esprit de corps because in this respect it will 
appear tyrannical. 

With this exception, the degree of correspondence is remarkable. Despite his own 
admission of empiricism in stating his principles, Fayol's thought is susceptible to 
schematic presentation. All the aspects of his theory of administration can be shown to 
correlate with each other, and to have their exact counterparts in Mooney and Reiley's 
scheme of the principles of organization. These facts are a testimony to the great 
weight of Fayol's authority in this field and further evidence of the essential unity and 
logic which underlie successful organization in practice. They constitute the strongest 
possible argument for intensifying the attempt to build up a recognized technique 
of organization and to secure that all who are likely to be responsible for administra- 
tion are trained systematically in this technique. 

Indeed the unique character of Fayol's work cannot be overemphasized. For 
the first time a successful business leader of long experience submitted, not the work 
of others, but his own duties and responsibilities to close scientific analysis. He viewed 
what he had to do as an administrator with a detachment as rare as it is valuable. 
In the first quarter century of the scientific study of business management, his is the 
only European figure worthy a place beside that of F. W. Taylor. To Taylor belongs 
the glory of the pioneer. He it was who initiated the idea that management and 
administration might be studied scientifically. But Fayol showed beyond question, 
what Taylor himself appreciated, but what many of his imitators have failed to em- 
phasize, that better management is not merely a question of improving the output of 
labor and the planning of subordinate units of organization, it is above all a matter of 
closer study and more administrative training for the men at the top. Seldom in 
history can two men working in an identical field have differed so sharply in methods 
and in the details of their careers and yet have produced work which was so essentially 

As was suggested in opening the lecture, the evolution of a science of administra- 
tion has a wider significance than the facilitation of business operations. It must 
have its repercussions in politics, in government, in every aspect of social life. It will 
change many things. But it is the author's conviction that only by developing such 


a science can we hope to arrive at any solution of the problems presented to us by 
the immensely rapid advance in and application of the sciences bearing on material 
things which have occurred during the last century. We must know more of social 
organization or a mechanized economy will run amuck with civilization. That is his 
excuse, if excuse is needed, for troubling you with a somewhat complicated piece 
of analysis. 







Dennison Manufacturing Company, Framingham, Mass. 

VOL. 26, APRIL, I932 




In the early days of factory management, when the problems and conditions 
were relatively simple, it fell to the lot of all sorts of human folk to manage the various 
jobs involved. Each used his own peculiar method and, naturally enough, it came into 
general belief that ability in management was an instinctive knack, that managers 
were born, not made, that few if any rules could be laid down, and that little could 
be learned by one from another. Even in the earliest days, a small handful of men 
called attention to the fact that management measures and forms of organization 
could be better or worse adapted to their uses; but for generations the suggestion 
passed unheeded. 

As the problems and conditions of factory management grew more complex 
and exacting, more men came to believe that the art of management was something 
more than an intuitive and highly personal knack. Before the Great War, a fair 
nucleus was beginning to study the art from the point of view of the forces involved, 
and with an eye to causes and effects. And the war experience of manufacturing 
strange materials, shifting conditions upside down, built up this nucleus into a very 
fair working minority. 

The first studies of management were largely descriptive. Forms of organization 
and management methods were described in detail, often with the result that those 
who read would try to take over the plans and the forms and the measures just as 
they were and apply them to their own tasks. This was the hey-day of the organization 
chart, when the anatomical structure of the organization for factory government was 
worshipped almost as if it were an end in itself. The troubles that arose when one 
tried to take over bodily the structure of one corporation or its card systems into a 
different problem at first set back the whole movement, and then forced studies 
which were more than merely descriptive. At the present time, we are beginning to 
accumulate fundamental material to form the basis of an art of management, which 
will be the application of the pertinent social and psychological sciences. 

Now a factory or business organization is, after all, one sort of social group — 
a group of human beings, each with his individual nature, attempting to work to- 
gether for some more or less definite end. A nation or a state is another sort of social 
group, living together and in no remote sense working together for common ends, 
whose management is the art we call governing. For the study of this art of governing 
I am proposing a shift of emphasis comparable to that which has taken place in 
studying the art of factory management, that is, a shift from the descriptive to the 
analytical; and from the analytical immediately to the engineering point of view, 
which focuses upon the natural material and psychological forces found in a given 
social group and the measures and structures of organization which can be applied 
to them in order to work toward its fundamental purposes. It is the approach, not 



of the historian or of the moralist, but of the student of applied science, the engineer. 

It will be granted at once that the problem is immensely more difficult than the 
problems which the nascent art of industrial management has to face; that in fact it 
is probably the most difficult of all human problems. It will be granted, also, that 
the sciences which this art of governing must learn to apply are themselves in a 
germinal, or at best embryonic, state. But two practical considerations outweigh any 
objections that can be raised. 

In the first place, few of the physical sciences advanced much beyond their 
embryonic stages until applications began to be attempted. Engineers fooled with the 
laws of physics and chemistry, and tried to make them work somehow, long before 
those laws were in any solid sense established; and these very attempts to make them 
work keyed up and directed the pure scientist in his further researches, so that he 
could elaborate or strengthen or amend these very laws. There is much in the history 
of the physical sciences to warrant the belief that they never could have grown to 
their present state unless the engineer had made successful and unsuccessful attempts 
at application, and had brought back the results of these attempts into the laboratory 
with a demand for further research. It can, therefore, be properly inferred that prog- 
ress in the social sciences will likewise depend upon earnest, even if sometimes unsuc- 
cessful, attempts at the application of what is so far known. 

Secondly, there stands the pressing need for progress in the art of governing, and 
especially in the art of self-governing. No human need today is greater. The very 
progress itself of the physical sciences and their application has plunged us into a 
world in which, if we cannot manage ourselves, we may flounder to our destruction. 
A benevolent and far-seeing czar of all the world might in fact wisely command prog- 
ress in the physical sciences and their application to stand still until progress in the 
social sciences and their application should catch up. There being no such czar, prog- 
ress in physical engineering will continue and so make necessary a still more rapid 
progress in social engineering. 

Four factors have in these modern days combined to bring this need for the study 
of government from an engineering point of view to the place which justifies the 
use of the word "overwhelming." First, the development of industrial and commercial 
processes, intensive and extensive, and the new complications and interrelations of 
all social and individual life which this development has brought about. Second, the 
steady reduction, since the great ages of discovery, in the proportion of new lands 
practicable for settlement — a reduction to a point today at which it is no longer a 
considerable factor in economic and social life. Third, the growth, intensive and 
extensive, of the deeply grounded social idea that the final purpose of all government 
must somehow include the satisfactions of all the people rather than the satisfactions 
of a god, a dynasty, or a small ruling group. And, fourth (and likely enough derived 
from the other three), the growth of an underlying demand for more general security, 
even at the cost of some reduction in the rate of progress in material well-being. 

These four factors are enough to warrant the belief that the greatest need of 
today, and of the next fifty years, is progress in the art of government. The present 
slogans of the business world which protest against any participation of government in 
economic affairs have many followers; but it is distinctly possible that the complexi- 
ties and the pervasiveness of economics will force every sovereign government into 


economic participation, economic influence, even sometimes into economic control. 
As a matter of fact, ever since the high tide of geographic pioneering, which turned 
somewhat before the end of the last century, the governments of the world have exerted 
more and more influence upon economic life; and this largely at the natural demand of 
first one and then another group of business men themselves. Taking into account this 
clear trend of the past, and the lessons, both positive and negative, of the great Russian 
and Italian experiments in the future, it is on the whole a safer guess that we will see 
from year to year more government in business, rather than less. 

But if a governing machine, built to fit the needs of a relatively simple agricul- 
tural and trading community, is loaded with the task of participation in an infinitely 
complex industrial structure, it is as certain that there will be a serious breakdown as 
that two and two make four. If the structure of the governing organization is to be re- 
garded as sacrosanct and cannot be altered to meet the rapidly changing needs of the 
modern community, it is true enough that we ought to have less government in busi- 
ness rather than more. But if the trends of the past are any guides to the future, we 
then find ourselves lost in a dilemma: Government both must and must not increase 
its participation in the economic life of this nation. 

The business answer to this dilemma has in recent years been that business must 
govern itself; that is to say, when put frankly, that the business world itself must build 
up a governing structure — a government. But when such propositions are examined 
in detail in any realistic fashion, problems of enforcement and of relative sovereignty 
arise which are insoluble. The political community and the economic community 
interweave and overlap to such an extent that they are indistinguishable. The gov- 
ernment of one is in fact the government of the other, and any attempt to bring down 
to cases a discussion of business self-government forces into clear view the fundamen- 
tally anomalous situation in which we find ourselves today. 

For it can be said in a very valid sense that the whole of society in the civilized 
world lives today as one society under two social systems — the economic system and 
the political system — between which there is practically no effective correlation ; 
two systems which many times become mutually incompatible. If, therefore, it should 
turn out that the business world can build for itself a government, it will be a wholly 
justifiable prediction that this business government will be more powerful than, and 
finally rise over, the political government; in which case we shall finally have a gov- 
ernment which not only participates in, but which is, the government of the business 
and economic world. 

But if the probability of this participation of political government in the economic 
world is discounted to the minimum, there still remains the job of the political world 
itself to cope with the complexities of modern society. In any case, the crucial need is 
for a study of the art of government, which regards the word "art" as a careful appli- 
cation of scientific knowledge and principles to the realistically analyzed forces of a 
given social situation. 


If political science is to furnish the basis for this engineering approach to the 
problems of government, it must view any community it is studying as a field of forces 
— psychological, biological, and physical. In view of what it can discover about these 


forces, it must determine the measures and structure of government which can be ex- 
pected to use these forces and to relate them so as to bring about development in the 
direction of the fundamental purpose adopted by and appropriate to a particular 
social group. 

At the very beginning, this fundamental purpose will need clarification, for the 
purposes of government have been in the past and are today various; and they are 
not, and have not very often been, clearly known or stated. There have been com- 
munities whose deepest purpose was the service of a god or gods, and others whose 
purpose was the satisfaction, wealth, power, or glory of a monarch. Some have aimed 
chiefly at the preservation of the status quo; others at defense against aggressive 
neighbors; some at the preservation or advance in wealth of an hereditary aristocracy 
or of the owners of property; and others, again, rather vaguely at the spiritual, mental, 
or material progress of the largest number of its people. Whatever the purpose may be, 
it is important to realize that its nature is from the point of view of an engineer one of 
the primary determinants of the measures of governing and of the forms of organization 
which are to be adopted. 

Of most concern in the modern world is the vague and ill-defined purpose in- 
cluded in what is often called the democratic ideal. When it is worked out in any defi- 
nite way, it seems to be a reaching for progress toward the fullest use and growth of 
each man's powers in the service of his fellows, and the enjoyment of such pleasures 
as contribute to the refreshment or enhancement of such powers. But it is still so 
vague as to serve much more often as camouflage for the narrow aims of a small group 
than as a determinant of government policies. 

It is with the true democratic ideal before him that a student of political science 

"It seems to me decidedly true that the time has come for a study of government from 
the following three angles: (1) government as the resultant of certain forces in the com- 
munity, which forces dictate (a) the direction and objectives of the activity of government, 
(b) the nature of its structure and its organization; (2) government as an organization for 
the attainment of certain objectives which are in part devised by government itself and are 
in part dictated to it from the outside; and (3) government as an agency having power to 
influence very strongly, and in some instances actually to create, the forces which in turn (a) 
influence the structure of government, (b) influence the objectives of government, (c) 
facilitate or impede the ability of government to attain its objectives." 1 

This way of specifying the needed study brings out at once the peculiar difficul- 
ties of the task due to the continuous interaction which takes place among the factors 
involved — the fact that causes do not stay causes, but are themselves altered and af- 
fected by their own effects — the principle which in the biological and social sciences 
has been called the principle of "circular response." These difficulties must be faced 
from the beginning, and will discourage all but the strong heart. They mean, cer- 
tainly, that while specialization in so complex a subject is essential, yet integration of 
the specialized studies must take place, as it were, simultaneously. The engineer in 
political science will find, as physicians have found, that the specialist is essential to 
progress but cannot be left to run the show alone. 

1 A letter from Professor John Dickinson. 



With some reasonably adequate idea of general objectives, the political science 
engineer must make progress in his understanding of the forces which must be dealt 
with; and here, particularly, is a realistic study necessary, for in the first place it is the 
forces in a given community, not abstract mankind-at-large forces, which are to 
be known. The stage of education and of culture, the peculiar conditions inculcated 
by "social inheritance," the special habits of mind and of vocation, make literally 
all the difference in the world. There are, of course, no two communities, no two 
nations, and no two states, principalities, or cities within any nation, in which the 
complex of forces is the same in their aggregate composition — just as there are no two 
business concerns in which the composition of forces is exactly the same, nor in any 
single community will they stay always the same, but will change with time and under 
the influence of such governing measures as are applied to them. The problem is in 
dynamics, not statics. 

The engineer will realize that the powers upon which the government of any 
community can draw are just those forces which actually exist among its membership. 
While these original forces may be enhanced, weakened, or redirected as time goes 
on, none can be counted upon at any given moment which is not then to be found 

A government must, therefore, find active among its members sufficient energy 
arising from biological, or from economic, political, or other social pressures, to serve 
its main purpose. This necessary energy may be very little, as in an isolated tribe 
located in a rich country, or very much, as in a nation with an itch for conquest; it 
may be sufficiently aroused by the pressures of physiological nature or of social 
tradition, with little interference of a political sort, or it may demand incessant and 
varied political activity. Whether simple or immeasurably difficult, the problem lies 
on the front doorstep of any government that, by whatever pressures may yield the 
most satisfactory net results, the energies sufficient to its main purpose be aroused 
among its members. 

But whether physiological, social, or political, these pressures are upon human 
beings, and upon just those human beings of whom the community is made up; and 
hence must be so adapted to the natures of just these men and women that there is a 
net yield of the required results. The practical problem of a government is to know as 
much as it can about the nature of its peoples and the influences actually at work 
among them, to judge when, where, how, and how much to interfere with the influ- 
ences of physiological nature or social custom then acting upon its members, and to 
decide to what extent to alter, lessen, or add to the pressures it is already exerting. 

The principal channels through which a government can bring its measures to 
bear upon a man are his desires to preserve his life, to avoid pain and discomfort, to 
mate, to preserve his family, to increase their standard of living, to enjoy companion- 
ship and social approval, to satisfy his curiosity and his moral or aesthetic demands. 
These desires and others which move men will practically never be free to find their 
satisfactions as they wish. They will be limited by each other, by physical nature, 
by the struggles of other men to fulfill their own desires, by traditions or unwritten 


laws, and by government. Just how much or how little any given government should 
spur, limit, or steer any of these desires is a practical question to be settled with refer- 
ence to the government's accomplishment of its fundamental purpose. In one case, 
this purpose may best be accomplished by leaving the control of men's activities to 
tradition or to their mutual interferences, with little affirmative governmental action; 
in another case, a different purpose — or even a similar purpose but in a different 
environment — may be served best by very broad and active government controls. 
No general principle can be laid down that for all governments and all peoples, under 
all circumstances, government control is per se good, bad, or indifferent. 

The measures which a government may use to spur, direct, or limit the attempts 
of its members to satisfy their desires will include propaganda or propagandist edu- 
cation, tariffs, subsidies, taxes and tax exemptions; it may make regulations as to 
mating and family life, working conditions, and the rights and duties pertaining to 

Many of the measures of a government must, of necessity, be negations and pro- 
hibitions. Naturally, where large numbers of corporations, institutions, or individuals 
are to be dealt with, the chief means of guidance must be negative, since it is more 
often possible to define the line of undesirable behavior than to select from among 
a large and rapidly developing variety of possible desirable behaviors the ones best 
for any individual case. In simple communities, it might be possible to prescribe a 
goodly bit of a man's life; in complex societies, it is more likely that the definitely 
undesirable areas will be proscribed, leaving open to individual choice a wide range 
of neutral or desirable action. 

But no regulation expressed in language can be clear in its application to all 
cases, even in a relatively simple and stable society; a series of interpretations of its 
application to a variety of cases may after a time give it practical clarification unless 
changes in conditions take place. In a changing community, judicial interpretations 
may be almost continuously necessary, leading, perhaps, to periodic alterations in 
the regulatory measures, each of which in its turn will need its own series of interpre- 

The limits of its appropriate activities are determined by the circumstances in 
which a government finds itself, and by its purposes. They may extend to the most 
intimate actions and thoughts of its members in a theocracy whose purpose is the serv- 
ice of a god or gods through obedience to a minute code; they may go almost as far 
under such a dictatorship as in Russia or Italy today; and so in close gradations the 
limits may narrow down nearly to zero, as in a moving frontier belt where each man's 
will and his power to exert it as against any other are the only limitations upon his 
acts. A people who have come by their early training to desire above all things peace 
and quiet and security will accept, and in fact demand, a close direction and regulation 
which would bring a restless folk to revolution. 

A government may be doing its best to ensure stability or to cultivate material 
progress; in most modern communities, it will be doing something toward both of 
these ends. As a result, unless there is a strong central planning and balancing power, 
a government will find some of its departments in fundamental conflict, as where the 
agricultural department urges restriction in planting while another department opens 


up new lands for cultivation. In this case, as in hundreds of others, some central 
clearing house with a persistent view to the fundamental purpose must have the 
facilities for balancing and integrating any differing or conflicting measures. In the 
simpler static communities, such integration is easy and rarely needed, but as com- 
munities grow in complexity, and especially where some sort of progress is part of 
their purpose, the need grows rapidly. 

For protection and stability, a government uses its police and courts, fire depart- 
ments, insurance, charities, an army of defense, and the like. For material progress, 
it may build roads, cultivate health, encourage invention, the expansion of business 
and industry, practical education and research. For spiritual progress, it may encour- 
age schools, the church, and the protection of what it considers moral, and help in the 
discouragement of what it considers non-moral attitudes and acts. In modern com- 
munities, non-governmental agencies and associations will be doing much of this 
work — for good or ill ; any government must take such organizations or efforts into 
account and decide to encourage, suppress, or let them run; it has responsibility com- 
mensurate with its actual power, whether it acts or fails to act. 

The total of education — by homes, schools, churches, papers and periodicals, 
unions, clubs, lecture halls, and political meetings — through personal contacts, 
radio, and movies, results in what is called a public opinion, which may be vague or 
concrete, mixed or simple, fixed or changeable, as the circumstances determine. As 
one of the forces to be taken into account by government, public opinion may be 
almost negligible, as in an illiterate and phlegmatic society governed by a powerful 
autocrat, or virtually self-determining, as among a group of energetic pioneer settlers. 
In either case, it can best be studied as a resultant of the influences which created it 
rather than as a thing in itself. 

The very foundation upon which public opinion at any moment builds itself is 
the body of beliefs, theories, prejudices, or superstitions with which a people have be- 
come indoctrinated early in life, and which have, therefore, gained the strength of 
habit. Those which result primarily in mental or emotional attitudes we may call its 
traditions; and those which are reinforced by physical activity, its customs. They are 
not unalterable any more than are other kinds of habits, but they can be changed only 
when conditions or influences conforming to the laws of habit-change are brought 
about or bring themselves about. 

The abstract validity of theories of government or of human rights — whether 
they be called moral, legal, or natural — is from a strictly engineering point of view 
wholly impossible to determine. For the government engineer, the only question of 
fact is what appeal the various theories make in any given social group — what force 
they exert or may exert, and, hence, what reactions follow them or may be expected . 
to result from them. For him, the classic question of the violability of contract by a 
sovereign is the question of the results to be expected in the group in question, if this 
or that sort of contract is violated. For him, the problems of the rights to liberty, prop- 
erty, or happiness, or the demands of loyal service, are to be subjected to a realistic 
analysis into the strength of their roots and the probable influence they will exert 
toward or away from the fundamental purposes of the government. If he is also a 
student of ethics, he will analyze them with the technique of ethics as well; but as 


government engineer he must focus upon them as sources of influence among the 
whole field of forces he has to survey. He will recognize that going theories, like the 
men who hold them, are the joint product of heredity and environment; immediate 
changes, therefore, are to be expected only through changes of circumstance or edu- 
cational influence. And customs, because in them mental habit is reinforced by 
physiological habit, he will recognize as even more powerful and difficult to change. 


It is with a view to the attainment of its fundamental purpose, and with an under- 
standing of the physical and psychological forces which it can call forth from the 
people, that the problems of governmental structures are to be approached. It is as 
true in government engineering as in any other branch of organization engineering 
that structure is most wisely to be considered as a means, not an end. But since its 
structure affects both the members of an organization and the effectiveness of its 
operating measures, it must, therefore, be determined and judged by its effectiveness, 
through its influences upon men and measures, in accomplishing the purposes the 
community holds. 

Not infrequently, an intense enthusiasm for a given structural form will fuse or 
confuse it with purpose; it is vital to government engineering that the two, if actually 
different, should be kept clearly separate, or if one and the same, should be so under- 
stood. For it is as easy for men whose real purpose may be the goal of the utilitarians, 
and who have seen a democratic form work their way in one case, to make demo- 
cratic forms their goal in all cases, as it is for aristocrats whose real purpose is to pre- 
serve aristocracy as a form to convince themselves that they are doing it for the "great- 
est good of the greatest number." Democracy, if thought of as a form of governmental 
structure, must submit to a practical inquiry as to whether it is the form for any given 
group through which the accomplishment of its purpose may be made most likely 
or least difficult. Democracy, however, is sometimes implicitly regarded, not so much 
as a form, but rather as the determination, for better or for worse as may happen, by 
the whole of a group of people of their own behavior. When this idea is made explicit 
and there is actually no other fundamental purpose but self-determination intended, 
then structure and purpose are more nearly one, and the main problems are as to 
what details of structure help most in the determination by a people of their own 
course of life. 

In any government there will be a number of different activities through which 
the main purpose is to be striven for; in a complex society, these will be many and 
varied. In any group, therefore, larger than a patriarchal household, more than one 
man will be needed, and work and responsibilities must be apportioned — the struc- 
ture of government must be departmentalized. To decide how this can best be done 
involves a listing of the tasks and analysis of the physical means and the human abili- 
ties demanded by each of them. But only in the simplest and most stable societies can 
these separate tasks, once provided for, be left without constant re-direction and co- 
ordination with each other. This demands their grouping into sub-sub-departments, 
and these into sub-departments and again into departments, which themselves must 
be related together in some one or another effective way. Moreover, as structure in- 


creases in complexity there is need of many cross relationships among many of its 
primary and secondary groupings which fall under different departments. 

Departmentalizing is so fundamentally necessary as to be practically universal 
in any governing task which is beyond the powers of one man. The main principles 
which direct its best use have, therefore, been worked out through wide experience; 
although in changing communities, with their changing and enlarging tasks and the 
growing differences in their inter-relationships, the reorganizations needed to make 
application of these principles take place very slowly. Much less thoroughly worked 
out are the methods and principles by which co-ordination among departments and 
sub-departments can be attained. This becomes especially necessary and especially 
difficult as a society comes to rely less and less upon physical compulsion and threats 
of compulsion, and as its governmental undertakings become more varied. In modern 
communities where economic projects, and where the facilities of communication, 
cross and recross those state lines and national lines which set bounds to the limits of 
government authorities, the needs for co-ordination have far outrun the development 
of its technique. 

Since any government structure is made up of men, men must somehow be 
selected for its tasks. The selection may be by birth — an eldest son, a youngest son, 
or a daughter — by tests of strength, of fighting power, by appointment, by examina- 
tion casual or elaborate, by lot, or by joint action of some designated people, that is, 
by election. The choice of method is a practical problem involving consideration of 
the nature of the place to be filled, the nature of the folk in the community, the exist- 
ing government structure and the purpose it serves. 

But, to be adequate, any method of selection must include a sufficient knowledge 
of the qualities and abilities the task calls for and the individual characteristics of the 
men from among whom selection is to be made. Some tasks, especially in modern 
communities, will need for their fulfillment a considerable degree of public support, 
and will require of the men who undertake them qualities which can call forth public 
acceptance; these will, therefore, be filled often by some more or less general election. 
The principles of government organization will demand, however, that such places be 
not complicated by adding to them tasks for which special technical abilities are 
needed. The guiding principles, indeed, of all departmentalizing, or specific task- 
setting, require that each job be determined upon, so far as is possible, in such a way 
that it will be reasonably likely that men able to do it can be drawn from the available 
material. For most of the separate tasks necessary to be performed in the management 
of a modern state, this is not difficult; the main trouble has been how to define the 
requirements of the job and analyze the characteristics of the possible candidates so 
that an adequate selection machinery might be set up. 

For the efficient performance of a task, as well as for the proper selection of its 
performer, a clear definition of it is essential; and not merely of the task conceived of 
separately, but of its relationships to other tasks as well. It is often easy to forget that 
an essential part of each separate task is to maintain all proper co-ordinate relations 
with other tasks. Any type of analysis or cutting up into parts seems to exert a fasci- 
nation which specializes and particularizes the mind, and so delays and interferes 
with the companion process of synthesis. 


Finally, if the structure of a government is to be fitted to its people and its pur- 
pose, it must take into account the probabilities of changes that may occur. In settled 
and relatively isolated societies, change may be very slow. But in newer communities 
and those where considerable changes in educational technique, in population, in 
economic environment, or inter-community relationship, take place, provision for ap- 
propriate steps in structural reorganization are essential. It is very rare that a govern- 
mental structure suited to one set of circumstances is well suited to any other. As early 
as 1901, Lord Bryce said: "Nobody now discusses the old problem of the Best Form 
of Government, because everybody now admits that the chief merit of any form is to 
be found in its suitability to the conditions and ideas of those among whom it pre- 
vails." 2 The penalty of undue stability must, therefore, be confusion, revolution, 
rebellion, or conquest. 

It is perfectly obvious that much fuller stores of knowledge than we now hold 
must be at our command before any application of political science fit to be called 
engineering can be made. Yet all the applied sciences have grown from inadequate 
beginnings by instant progressive attempts to make useful application of whatever is 
known. It is these attempts and their inadequacies and partial failures which have 
given impulse and direction to the search for the needed facts and to the experiments 
in method. It seems impossible to develop an engineering art otherwise. It is none too 
early, then, for the beginnings to be made in the application of scientific knowledge 
and method to the art of governing — in the development of an objective and real- 
istic Political Science Engineering. 

2 " Studies in History and Jurisprudence," Oxford Press, 1901. 



From the Graduate School of Business Administration, Harvard University, Boston, Mass. 


XVIII, NO. 7, SEPTEMBER, 1 936. 



The human aspect of modern industrial organization attracted no great attention 
from industrialists until the advent of the European War of 1914; after the outbreak 
of hostilities the need to accelerate and to maintain production of various necessities 
quickly provoked a formulation of the problem. Apparently no one had ever suffi- 
ciently considered the enormous demand upon industry that would be made by a 
war-machine organized upon so heroic a scale, for armies counted in millions were a 
gigantic innovation. Nor had anyone considered the effect of the strenuous and sus- 
tained exertion imposed upon those who worked to provide supplies. The authorities 
in England speedily became aware of a "national lack of knowledge of the primary 
laws governing human efficiency." In particular, there was "need for scientific study 
of the hours of work and other conditions of labor likely to produce the maximum 
output at which the effort of the whole people was aimed." The actual conditions 
of work set in the munition factories were, in the early days of the struggle, admitted 
to be progressively detrimental to the worker and consequently unfavorable to that 
maintenance of output for long periods upon which success in large part depended. 
Inquiries undertaken by Dr. H. M. Vernon and others appointed by the British 
Health of Munition Workers Committee were effective; the introduction of shorter hours 
of work, of rest pauses and of other humane innovations resulted in the restoration of 
morale and the maintenance of production. 

The stimulus to inquiry provided by this situation has continued to operate in 
the years that have elapsed since 1918, and much has been learned of the human fac- 
tors that are involved in industrial organization for work. Another paper given at 
this symposium, by Dr. D. B. Dill, will present some of the most recent and inter- 
esting findings by physiology in the problem of fatigue. In addition to physiology, 
psychology has been summoned by industry to aid its determination of appropriate 
working conditions for the human being in action. In many of these inquiries, and 
rightly, the worker is considered and studied as an individual. Up to the present there 
have been few attempts to study the social conditions of work — that is to say, the 
effect upon morale and production of the situation created by the interrelation of 
several human beings in an industrial department. This paper is designed to give a 
brief account of two experiments at the Hawthorne Plant of the Western Electric 
Company which in a sense stumbled upon this problem as an exceedingly important 
aspect of industrial administration. 

Officers of the Western Electric Company were probably moved to begin the 
researches of 1927 at Hawthorne by two chief influences. The one may be described 
as a general interest in the problems of the incidence of "fatigue" and "monotony" 
in factory work. This took its rise, remotely, in the publication by the English Indus- 
trial Health Board of its studies of the conditions of work in munitions factories and 
other industrial plants during and after the War. The other influence may be at- 
tributed to the fact that, as good engineers, these same officers were aware that 
Company policies with respect to human beings were not so securely based as policies 



with respect to materials and machines. Mechanical processes, the type and quality of 
materials used, were based upon carefully contrived experiment and knowledge; 
the human policies of the Company, upon executive conceptions and traditional 
practices. In determining the human policies, the Company had no satisfactory cri- 
terion of the actual value of its methods of dealing with people. A mechanical process 
as studied in a modern factory will in some way reveal an inefficiency; a traditional 
method of handling human situations will rarely reveal that it is rooted in mere use 
and custom rather than knowledge. The hope that an experimental method of assess- 
ing the human effect of different working conditions might be discovered must be 
regarded as counting for something in the development of the experiment. The object 
was not to increase production but to discover facts. Where production could be ac- 
curately observed, records were kept; but this was arranged merely as an essential 
part of the experimental procedure. 

The original "test room" was installed in April, 1927, and continued to operate 
until 1932, when, for reasons not connected with the experiment, it was abandoned. 
The principal results here presented have been described by T. N. Whitehead. The 
operation selected for study was that of the assembly of telephone relays. Five girl 
workers were transferred from their usual work surroundings into an experimental 
room, within which their work was supervised by specially appointed observers. 
Arrangements were made for the continuous and accurate record of individual out- 
put and for other observations that were considered appropriate. It will suffice for 
our present purpose if we say that the output of the test room workers continued to 
rise slowly for a period of years. In its upward passage this major gain ignored almost 
completely experimental changes arbitrarily introduced from time to time by the 
officers in charge of the experiment, including modification of the hours of work, the 
introduction of rest periods, etc., except those that modified the social organization 
of the group. 

Thus it became evident that the group was performing two distinguishable, but 
mutually dependent, functions. On the one hand, the group was performing its 
technological, or economic, function; on the other hand, the activities of the group 
were being so modified and controlled as to heighten the social solidarity of the group 
within itself, and also to stabilize its relations with other groups. 

As time went on, the interplay between the economic and the social functions 
of industrial groups became the central subject of the Western Electric researches. 
Although the girls' work itself was not altered, the general conditions in the test room 
differed in a number of respects from those to which the girls had previously been 
accustomed. They were informed as to the nature of the experiment, and their co- 
operation was invited. The girls were paid, as before, on the same system of group 
piece work; but their previous group had contained over a hundred individuals, while 
in the test room the group, for purposes of payment, consisted of only five members. 
They were told to work at a comfortable pace, and no emphasis was laid on achieving 
any given level of output; in fact, they were warned against "racing" or forced output 
in any form. Conversation was permitted; on occasion it was general, while at other 
times it was confined to pairs of neighbors. In addition to these innovations, certain 
other experimental changes were introduced from time to time. The length of the 



working day was varied, as were also the number of working days in the week, rest 
pauses were introduced, and so forth. 

Figure 1 is a photograph of the girls at work. They are sitting in a row along their 
work bench, and opposite them are trays containing the necessary parts for assembly. 
Immediately in front of each girl is a small jig to assist her in building the relays. 

Numerous records were kept by the supervisor in charge. Since an automatic 
instrument recorded, to a fraction of a second, the instant at which each girl com- 
pleted every assembled relay, we have the minute-to-minute output of each indi- 
vidual over a period of five years. Other records relate to quality of output; reasons 
for temporary stops; the length of time spent in bed every night; periodical medical 
reports, and so on. Room temperatures and relative humidities were taken hourly 
and the Chicago Weather 

Bureau has supplied simi- 
lar records for outside 
conditions. In addition to 
these, a number of other 
records were kept. 

The supervisor and 
his assistants made ex- 
tensive daily notes of 
conversation, and of the 
relations developing be- 
tween the workers. The 
workers were also sepa- 
rately interviewed by an 
experienced person on a 
number of occasions in 
another room. In all, we 
have volumes of contem- 
porary observations bear- 
ing on the characters and 
dispositions of these girls, their mutual relations, their conversations and attitudes, 
their home situations, and their leisure hour acquaintances and activities. All this 
information was collected with the knowledge and consent of the girls themselves; 
and, as evidence that this consent was real, we may add that two suggested records 
were vetoed by the girls and so were never kept. 

Figure 2 shows the weekly output rates for each of the girls in the test room. 
Each point, or short horizontal bar, indicates the average number of relays assembled 
per hour by the worker for the given week. 

It will be noticed that girls Nos. 1 and 2 did not enter the test room until the 
beginning of 1928, when they replaced the original girls 1A and 2A. Similarly, No. 
5 left the Company in the middle of 1929, and returned about 10 months later. Dur- 
ing this interval her place was taken by No. 5A. In this short account no further men- 
tion will be made of Nos. 1 A and 2A, but No. 5A entered into the life of the group 
too vitally to be ignored, and she will be referred to in due course. 

Fig. 1. — Girls at bench assembling relays. (Thanks for reproducing these 
illustrations is due the editors of Human Factor.) 



All these girls were experts at relay assembly, having already had several years' 
experience, but it is noticeable from the figure that in every case their output rate 
increased substantially, the average increase being somewhere in the neighborhood of 
30 per cent. 

Of great interest are the wave-like irregularities exhibited by each graph. Some 
of these "waves" last for months, and others only for a week or two. Moreover, the 






• ora CKW6C tua 



1l \F 


-Jv ' ' " " 

•|, r 











-1 — 


ATM f. 



— i — 






Ik 1 

^ ^ V G 




TeWT 3 










OttRATOR NO 3 i 






■* B 



] tV* 

** : *l7ln 


tA+ **—■£*■ 



/W 1 * 





l^ 4 *— - 

— - u- 



u ~v 




w* 1 













n .n 





n F 


v fi 




L ' " 


nf p-ij l 





y L 

v \ 

(Ur " 








operator * 


1 o> S 


,n C 

j 1 J3y> 

1*1 n 

, r. 

j\H ^ 


—i ' ^iF 

" L? ' 







jjAr 1 * 






Nil i 1 



I io 1 

II 1 u. 



li. 1 




1 18 

1 w 


■o p 

| B n 




JO b 


J» M« Hff 

i- ^ li' [K Jsn Mx K) Ui Sol 

b * » * * ;i n ■ a 

KT30 I l<?3l 


Fig. 2. — Weekly output rates for girls in test room. Each point or short horizontal bar indi- 
cates the average number of relays assembled per hour by the worker for the given week. 

available output figures show that similar irregularities occurred with durations of as 
little as a minute or two upwards. At first it was supposed that these variations in 
working speed might be related to the experimental changes deliberately introduced, 
or possibly to other changes in physical circumstance, such, for instance, as tempera- 
ture, or the worker's own physical state. But a careful analysis of the data forced us 


to the surprising conclusion that irregularity in speed of work substantially failed to 
correlate with any changes of physical circumstance. This applied equally whether 
experimental changes, irregular changes of natural circumstances, or those cyclical 
changes involved in the passage of time were considered. 

Since working speed was in fact very variable, and yet so insensitive to changes 
in physical circumstance, changes in the girls' social relations were next examined as 
being possibly connected with variations in working speed. And this time positive 
results were obtained. Speed of work varied markedly with changes in the sentiments 
entertained by the workers towards each other, towards their supervisors, and to- 
wards the group as such. To give a social history of the test room from 1927 to 1932 
would be to give an explanation of the major trends shown in these graphs. 

We have already remarked that these graphs show an average increase of speed 
of about 30 per cent. In any ordinary sense they are not learning curves, for the work- 
ers all had several years' experience in this particular work, and they had reached a 
more or less steady state as regards speed and skill. 

Nevertheless, the plateaux and spurts are decidedly suggestive. It was the or- 
ganization of human relations, rather than the organization of technics, which ac- 
companied spurts in these cases. This illustrates the futility of attending exclusively 
to the economic motivation of workers, or to their physical conditions of work. These 
things are of high importance; but no group of workers can be expected to remain 
satisfied, or to co-operate effectively unless their social organization and sentiments 
are also protected. 

In looking at Figure 2 once more, each graph can be imagined as the resultant 
of a number of wave-like disturbances of differing wave-lengths each superimposed 
upon the other. An ocean, such as the North Atlantic, only too frequently provides 
us with an analogy. So often it is simultaneously disturbed from shore to shore by 
ripples, waves, oceanic rollers, and by tides — four different types of wave-like dis- 
turbances, differently produced, and each with its characteristic range of wave- 
lengths. In similar fashion a work speed graph can be thought of as simultaneously 
disturbed by superimposed fluctuations of very different wave-lengths, or better, 
time-spans, for in this case the length of the wave is measured in units of time. 

By means of a statistical device it is not difficult to break down a work speed 
graph into a family of curves, each curve containing only those fluctuations lying 
between certain limits of wave-length, or time-span, and such that, if the ordinates 
of all these curves be added together, the original graph results. This has nothing to 
do with harmonic analysis; it is just a device for examining all disturbances of ap- 
proximately similar time-spans in isolation from the others. 

This analysis has been performed by Whitehead for each of the work speed 
graphs in Figure 2. For instance, among others, one curve has been prepared for 
each girl, showing all those fluctuations of speed whose time-spans lie between 1 and 
4 weeks — all other fluctuations have been eliminated from these particular curves. 
And these 1-4 week time-span curves have been compared with one another to see 
in what degree the speed fluctuations of one girl correspond with those of another. 

The accompanying figures give this information for typical dates throughout the 
experiment. Figure 3 refers to April, 1928. The five circles represent the five relay 






Fi£.S April m& 


Fi£.S October 1929 

Fig. 4 July 1929 

assemblers in the order in which they sat, the distinguishing number of each girl being 
placed within her circle. When the speed fluctuations of two girls show a significant 
correspondence in the 1-4 weeks time-span, then the pair in question are joined by a 
line; for example, the pairs 1-2 and 3-4. The figures against the lines indicate the 
strength of the correspondence stated in percentages, and when this correspondence 
amounts to 50 per cent or more this is indicated by a thick line. These figures do not 
refer to correlations (V), but to the squares of correlations (r 2 ), and are called "de- 

Determinations have certain theoretical and practical advantages as compared 
with correlations; but the point to remember is that a determination is always nu- 
merically smaller than its corresponding correlation. Thus a correlation of 0.8, or 
80 per cent, corresponds to a determination of 0.8 2 , or 64 per cent. All the determina- 
tions shown in the following figures are positive unless otherwise indicated. 

With this preliminary explanation let us ex- 
amine Figure 3. The relay test room had been 
running for about a year, but Nos. 1 and 2 had 
been members of the group only for the last 3 
months of that time. As the figure shows, these 
two girls "determinated" quite strongly. No. 2, of 
Italian origin, was undoubtedly the leading mem- 
ber of the group. She was the fastest worker, 
showed the highest score in an intelligence test, 
and possessed the most forceful character in the 
room. This girl was ambitious and at times hoped 
to obtain a secretarial post, but circumstances had 
prevented this. At the date in question, April, 
1 928, her mother and sister had recently died, and 
No. 2 ran her father's house, looked after her 
younger brothers, and was the principal wage- 
earner of the family. In the main shop, No. 2 had found little scope for satisfaction, 
but the test room seemed to offer a greater outlet for her ambitions and energies, and 
she threw herself into the new situation with vigor. Her friend, No. 1, of Polish origin, 
possessed a more placid disposition and contentedly followed her strong-minded neigh- 
bor. This friendship lasted throughout the five years of the experiment. 

Workers 3 and 4, both of Polish origin, had been friendly since the beginning of 
the experiment, although they had little in common except their chance propinquity. 
Neither possessed conspicuous qualities of leadership, though it appears that, in the 
absence of No. 2, No. 4 might possibly have assumed a dominant position in the group. 
Characteristically, No. 5 shows no correspondence in her speed fluctuations with 
any of the others. A Norwegian by birth, she had lived in the States only a few years 
and spoke English with difficulty; and this, combined with the fact that she was 
married, older than the others, and of a phlegmatic disposition, prevented her from 
ever entering fully into the life of the group. 

So Figure 3 shows determinations between two pairs of friends, but nothing that 
could be described as a general state of mutual influence as between the five girls. 

Fig. 6 January 1930 

Figs. 3-6. 

-Correspondence in output 


Figure 4 illustrates the state of affairs fifteen months later, in July, 1929. Here 
every girl "determinates" significantly with every other girl, and in many cases the 
degree of determination is decidedly great. It will be noticed that every determina- 
tion exceeds 50 per cent, except those involving No. 5 which never exceed 30 per cent. 

During the fifteen months separating Figures 3 and 4, the group had been ac- 
quiring common activities, interests and loyalties. To a large extent the group had 
taken their discipline out of the hands of the supervisor and were performing this 
function for themselves. To give only one instance: when a girl wished for a half day's 
leave of absence she had to obtain permission from the supervisor. But a custom had 
become established by which no girl could ask for such permission unless the group 
sanctioned it. This leave was seriously debated by the group and not always granted, 
and we do not think the supervisor ever reversed the decision of the group in this 
matter. Group solidarity had developed with No. 2 as the unofficial but acknowledged 
leader; and Figure 4 shows the extent to which the girls were influencing each other 
in respect to speed fluctuations in a particular time-span. 

Figure 5 shows an almost complete collapse of group solidarity only three months 
later, in October, 1929. In the interval between Figures 4 and 5, No. 5 had left the 
Company's employment of her own free will, and had been replaced by another relay 
assembler, No. 5A. This last girl was selected at the request of No. 2, the two having 
been close friends for some years. 

No. 5 A was in all respects more congenial to the group than her predecessor; 
she was of the same age as the others, unmarried, good-tempered and co-operative. 
Nevertheless, No. 5A was unaccustomed to working with her new associates and quite 
unused to the type of integrated group activity of which she had no previous experi- 
ence. The result was that, in the 1-4 weeks time-span, she "determinated" with no 
one. But, even more significant, other pairs not involving No. 5A also failed to show 
much mutual influence. Solidarity had fallen all round in consequence of an un- 
assimilated social element, and this in spite of the fact that No. 5A was more popular 
than her predecessor. 

However, this state of affairs did not last long, and three months later (Figure 6) 
we find the highest state of integration ever observed in the group in this particular 
respect. No. 5A had established herself as a participating member of the group and 
being better adjusted in her surroundings than No. 5, determinations are higher. 
Every individual determinates with every other; the mean value of all the determina- 
tions is 64; and the least is 37, corresponding to a correlation coefficient of 0.6. 

Taking Figures 4 and 6, the only cases here presented of complete determinations, 
the mean values of the determinations for neighbors are 66; for two individuals sepa- 
rated by one other, 57; for those separated by two others, 47; for those separated by 
three others, 43. 

Figure 7 shows another change by June, 1930, five months later. Two events 
occurred to account for this. In the first place the girls had changed seats, as can be 
seen from the figure. 

This rearrangement was introduced for experimental reasons in April of 1930, 
and was maintained for ten months. A change in seating may not seem important to 
those whose occupation permits them some liberty of bodily action during working 



hours, but it made a great difference to the relay group. Their work necessitated 
some degree of visual attention, as well as continuous finger and arm action; it thus 
determined the position of the body with respect to the work-bench. So, of necessity, 
intimate conversation could only be carried on between neighbors, although general 
conversation in a raised tone of voice was not uncommon. Thus a change of seating 
order involved new associates, an entirely different perspective of the whole group 
and consequently the need for a new orientation. 

Shortly after this change No. 5 requested the Company to take her back and to 
place her again in the test room. The circumstances surrounding this request were 
pathetic and the request was granted. But the responsible official had not realized 
the extent to which this action would be resented by the remainder of the group. 
The fact was that No. 5A was decidedly popular and was supporting an invalid father. 

Her removal from the test room resulted in a small 

drop in her wages (she was given a somewhat 

F£7 June 1930 

Fl£ 8 September IBM 

^ i- eS 

Fig. 9 March IS3I 


Fig. 10 January B51 

Figs. 7-10 



F~q I I* ^ lz different occupation), and it was supposed to in- 

crease her chance of being laid off as a result of the 

Q " l6 q — industrial depression, which was then in its early 

J=0 — I I__zz0 ' stages. For both these reasons the reinstatement 

of No. 5 was disliked by the group, and this dislike 
was transferred to No. 5 herself. In Figure 7 No. 5 
has no relations with anyone, and other determina- 
tions are few and relatively weak. However, by 
September of 1930, three months later, the group 
began to show signs of reintegration (Figure 8). 
How far this might have gone it is impossible to 
say, for shortly afterwards the girls were put back 
into their original order of seating. The result is 
shown in Figure 9. Not only has integration in 
this respect collapsed, but Nos. 2 and 5 show a marked negative determination (the 
corresponding correlation equals 0.72). It will be remembered that No. 5A was a 
close friend of No. 2, and at this time the latter rarely spoke to No. 5 except to snub 
her. The result is seen in the variations of their working speeds; when one worked 
faster the other worked slower. Their speeds varied in anti-phase. This is the only in- 
stance of a negative correlation in the 1-4 weeks' time-span, though more frequent 
instances are found in some of the other spans. 

The occurrence of this one negative determination involving No. 2 should per- 
haps be considered in connection with the fact that the total of her determinations 
contributed about 40 per cent more than their proportional share to the total of all the 
determinations of Figures 3-10. In other words, determinations in which No. 2 is 
involved are approximately twice as great as those in which she is not involved. This 
presumably is the mark of her leadership, and it is possible that the one negative 
determination may be partly due to the same influence. 

Finally, Figure 10 shows the situation about ten months later, in January, 1932. 
By this time the depression was at its height, many employees had been laid off and 
quite evidently the process was bound to continue. For this reason, as well as others, 


the tone of the relay group had been gradually changing from optimism to resignation. 
The future was a matter for dread rather than hope, and No. 2 in particular was 
again restless in the feeling that her ambitions were not being realized. The self-disci- 
pline of the group showed signs of deteriorating, and the ten months separating Fig- 
ures 9 and 10 resulted in little growth of solidarity. The group had lost its spring. No. 
5 was no longer the object of active resentment, but she was effectively outside the 
common life and practically never spoke to any of the other girls. 

We have briefly compared synchronization of fluctuations, in the 1-4 weeks time- 
span, with the social sentiments of the workers, themselves; and it appears that these 
two factors tend to vary in close accord with one another. This accord was consider- 
ably more detailed than has been explained. Moreover, the story is much the same 
whatever time-span be considered. For fluctuations have been examined with individ- 
ual endurances of a minute or two, at one end of the scale, up to those having endur- 
ances of nearly three months at the other extreme. 

In every case, the degree in which the speed fluctuations of two workers corre- 
spond relates itself in some fashion to their mutual sentiments; though the type of social 
sentiment involved does depend somewhat on the length of the time-span chosen. 

The second experiment we report has been fully described in a monograph en- 
titled "Management and the Worker," by F. J. Roethlisberger and W. J. Dickson. A 
study of a bank-wiring room, it constituted the final experimental phase at the end 
of five years' inquiry at the Hawthorne plant. "The method of study was novel in that 
it utilized two types of investigation simultaneously. One type consisted of an indirect 
conversational interview, the other of direct observation. The interviewer remained 
as much as possible an outsider to the group, and the interviews were held by appoint- 
ment in a private office. The observer was stationed with the group in the role of a dis- 
interested spectator. His function was to keep records of performance, and of events 
and conversations which he considered significant. . . . The attention of both in- 
vestigators was fixed upon the same group, and the one simply attempted to get infor- 
mation which the other could not get as well, or could not get at all." 1 

The situation that was revealed in this method of study was very different from 
that which obtained in the relay assembly test room. Whereas in the test room the col- 
laboration of workers with management was at a high level and the aims of the two 
groups apparently in perfect accord, the situation in the bank-wiring observation 
room was quite otherwise. The spontaneous social organization of the latter group 
seemed to order itself about a certain sentiment of fear or doubt with respect to 
managerial intention. The group had arrived at a conception of a day's work which 
was less than the "bogey" set for the department. This decision was not the result of 
careful consideration or logical process; it had "just happened." And the output rec- 
ords showed a "straight-line" production which was closely in accord with the group 
conception of a day's work. There were a variety of methods by which group discipline 
was enforced; no individual escaped the group decree. As a result of these practices, 
the departmental efficiency records did not reflect the actual situation. The official 
incentive plan was not functioning as it was intended. The group had so organized 

W. J. Dickson, "Actual Behavior in a Shop." Paper read to Personnel Research Foundation, January 25, 1935. 


itself — in a purely spontaneous fashion — that the intentions of the engineer organiz- 
ers were defeated. 

To show how completely the informal organization of the group had defeated the 
official plan, we quote certain findings made by W. J. Dickson, the officer in charge of 
the investigation. In the original test room experiment, study of the performances of 
the girl workers after a period of years showed that the comparative achievements of 
the individuals in respect of output gave a ranking order that coincided very closely 
with the order assigned by intelligence and vocational tests. Quite a contrary situation 
revealed itself in the bank-wiring room. The relative rank in output of the various 
workers was compared with their relative rank in capacity as measured by tests of 
intelligence and dexterity. "This comparison showed that there was no relation be- 
tween their ability as measured by these tests and their actual performance. The low- 
est producer ranked first in intelligence and third in the dexterity tests. The man who 
scored highest in the dexterity test ranked seventh in intelligence and seventh in output. 
The man who scored lowest in the dexterity test shared first place in intelligence and 
ranked fifth in output. This then was a situation in which the native capacities of the 
men were not finding expression in their work. In order to see whether differences in 
earnings accounted for difference in output, these two factors were then compared. 
Here again no relation was found. The man who ranked first in earnings ranked 
fourth in output, and the man who ranked lowest in earnings ranked fifth in output. 
Two of the men received the same wages, yet one produced an average of 1 6 per cent 
more work." 2 Dickson goes on to point out that careful analysis of all the available 
data showed that these differences in output related themselves approximately "to the 
individual's position in the group." That is to say, differences in output related them- 
selves to the social controls established by the informal grouping and not to individual 
capacity or to economic or logical considerations. Dickson's conclusions, briefly sum- 
marized, follow. He expresses a caution that these conclusions apply specifically to 
the group under observation and are not to be interpreted as generalizations. 

1 . Output for these operators was a form of social behavior. Their own peculiar 
concept of or feeling about a day's work was the idea about which the informal 
grouping was organized. Individual differences of performance were related rather 
to the individual's position in the group than to his actual capacity. 

2. In this situation the supervisory controls established by management had 
failed. The whole logic of the technical organization was implicated in this failure. 

3. The problems encountered were not due to a logical insufficiency in the wage 
plan. Such plans assume that a worker is primarily moved by economic interest and 
that he will act logically. The study shows that social considerations outweigh the 
economic and logical; the workers' actions were based upon non-logic and sentiment 
— in other words, upon social routine procedures. 

4. The study has some interest for vocational work in selection. In the conditions 
described, scientific selection and placement do not ensure a corresponding efficiency. 

5. The conflict of loyalties involved in the dissonance between the official and 
the actual organization leads to worry and strain, not only for the supervisor but also 

- W. J. Dickson, op. cil. 


for the workers. There is reason to believe that the girl workers in the test room, whose 
work closely reflected their capacities, experienced a feeling of release from this type 
of strain. 

Roethlisberger and Dickson have elsewhere pointed out 3 that situations such 
as that described are commonly misinterpreted. It is supposed, for example, that 
there is a necessary hostility somewhere implicit in the relations between management 
and the worker, and that restriction of output is consciously contrived. But this type of 
explanation misses "the essentially non-logical character" of the situation. It is only 
those who assume that economic interests dominate the individual and that clear 
logical thinking serves these interests who can argue thus. And it is the assumption, 
and not the fact, that drives them in this direction. "Logically it could be argued, for 
example, that it was to the economic interest of each worker to produce as much as 
he could and to see that every member of his group did the same. But, in actuality, the 
the workers saw to it that no one's output ever exceeded a certain limit. Had they been 
asked why they limited their output in this way, they would probably have expressed 
fears of a possible 'rate reduction.' Yet none of the men in the bank-wiring observation room 
had ever experienced a reduction of rates. But they acted as though they had. This behavior was 
not directly in line with their economic interests, nor was it based on the facts of their 
own experience with the company." 

There are other observations which make the hypothesis of a necessary hostility 
between management and workers untenable. The Western Electric Company has a 
long and consistent record of fair dealing with its employees. Both overtly in a very low 
labor turnover and verbally in the interviewing program, employees have shown a 
manifest appreciation of the company's attitude. "In the interviews of 1929, when over 
40,000 complaints were voiced, there was not a single unfavorable comment expressed 
about the company in general." It seems to be clear that the hostility hypothesis is an 
inference from the insufficient assumption of economic interest and logical thinking 
as fundamentals of human association; it is not an inference from the facts. 

"Upon examining more closely the behavior of employees," it became evident 
that "many of their actions were of the nature of mechanisms to resist too rapid changes 
in their environment." The opposition to change was not only reflected in all their 
tactics to keep output constant, but was also implied in all the reasons they gave in 
justification of their actions. Roethlisberger and Dickson, therefore, conclude that the 
chief function of the informal groupings which organize themselves on the working 
line is "resistance to change or any threat of change in their established routines of 
work or of personal interrelation." This is a very ancient and well-known character 
of human association. 

Scientific management has preferred to work with explanations or hypotheses 
that are simple and logical. This would be admirable, if one were not guilty of what 
A. N. Whitehead terms "the fallacy of misplaced concreteness." * But no study of 
human situations which fails to take account of the non-logical social routines can 
hope for practical success. Roethlisberger and Dickson give two reasons for the failure 

3 " Management and the Worker, Technical vs. Social Organization in an Industrial Plant." Harvard Business 
School, Bureau of Business Research, Boston, 1934. 

4 A. N. Whitehead, "Science and the Modern World," The Macmillan Company, New York, 1926. 


of the technical plan in the bank-wiring department. They begin by distinguishing 
between the technical organization and the informal social organization of the em- 
ployees. The technical organization is chiefly remarkable "for its logically contrived 
character" which makes for rapid change. And, just as capacity for change character- 
izes the technical organization, so resistance to change characterizes the informal em- 
ployee groups. As a result the employee groups feel that they are constantly under 
fire — in the sense that it is as if the technical experts were constantly battering to 
pieces any routines of collaboration that the workers develop. This rouses resistance, 
first, because "the worker is at the bottom level of a highly stratified hierarchy. He is 
always in the position of having to accommodate himself to changes which he does not initiate." 
Second, "many of the changes to which he is asked to adjust rob him of the very 
things that give meaning and significance to his work. His established routines of 
work, his personal relations with fellow-workers, even the remnants of a cultural 
tradition of craftsmanship — all these are at the mercy of the technical specialist. He 
is not allowed either to retain his former traditions and routines or to evolve new 
ones of any probable duration. Now, the social codes which define a worker's relation 
to his work and to his fellows are not capable of rapid change. They are developed 
slowly and over long periods of time. They are not the product of logic, but of actual 
human association; they are based on deep-rooted human sentiments. Constant inter- 
ference with such codes is bound to lead to feelings of frustration, to irrational exasper- 
ation with technical change in any form." Ultimately, therefore, the disrupted codes 
revenge themselves by giving rise to informal employee organizations which are op- 
posed to the technical authority. 

Scientific management has developed a logical flexibility which in itself is ad- 
mirable. As applied to industrial organization, however, this flexibility perhaps cre- 
ates more problems than it solves; it is at least true that social sentiments and 
routines of association find difficulty in adjusting themselves to the rapid "shift." But 
scientific management has never studied the facts of human social organization; it has 
accepted the nineteenth-century economic dictum that economic interest and logical 
capacity are the basis of the social order. It would seem possible, therefore, that scien- 
tific management has itself done much to provoke that hostility between management 
and workers which now so inconveniently hampers every development towards 

There remains comfort, however, to be drawn from the series of experiments at 
Hawthorne. These experiments: 

1 . Have called attention to an important group of facts — the facts of spontane- 
ous social organization at the working bench. These facts are of such a nature that 
they escape the notice of physiological and psychological inquiry. 

2. The experiments strongly suggest that the excellent results obtained in the 
original test room were largely due to the achievement of a comfortable equilibrium 
between the technical organization, or plant authority, and the spontaneous social 
organization of the workers themselves. This equilibrium is not generally found in 
working departments or in other factories. 

The facts that we have been considering have revealed and have led us to describe 
certain characteristics of two social systems, one experimentally (though accidentally) 


produced, the other a spontaneous formation whose existence had been ignored. Now 
there can be little doubt that any group of people who work together sooner or later 
take on some of the characteristics of such a system, and thereafter so act that their 
behavior can only be conceived as the resultant of social forces as well as of economic 
forces and of those psychological forces that are private to the individuals. Social 
organization is, in fact, a human need; it is, in some measure, necessary and inevitable. 
Its mere existence disciplines the members and gives rise to sentiments, often very 
strong sentiments, of loyalty, of personal and group integrity, and not infrequently 
of pride. No. 2 of the test room, commenting after the event on the decline of output 
toward the end of the long experiment, at a time when the effects of the depression 
were at their worst, remarked, "We lost our pride." 

Not infrequently the proprieties of the spontaneous social system call for more 
consideration than the strictly economic interests of the group or the psychological 
properties of the individual members of the system. This is because the social forces 
are nearly always strong, and sometimes dominant. Like that of the family, the social 
organization of any group is felt as a real thing, indeed as something far more real 
than the technical organization of a factory; and spontaneously formed human rela- 
tions are felt to have a meaning and a value that are lacking in purely hierarchical 
relations or in those relations that are involved in merely working together in time 
and place, according to an arbitrarily determined plan. 

These statements, assuredly, are platitudes; but they are platitudes that are 
nearly always forgotten by most men of action in action. Yet, it is urgently necessary 
that they should be always remembered, — remembered and acted upon. Whenever 
men work together search should be made to discover and to characterize the social' 
systems that have spontaneously arisen, for these are the worlds in which the individual 
members feel themselves to be living, just as a mother feels herself to be living not so 
much in a particular house, in a particular street, in a particular town, as in her 

If you are talking with, giving orders to, planning for, making use of a man who 
is living in a world — or feels that he is living in a world — of which you are ignorant, 
whose existence you do not suspect, is it possible that what you say to him, what you 
command him to do, what you plan for him, what you use him for, will have for 
him the same meaning that it has for you? Is it not more probable that what seems to 
you good for him will to him sometimes seem bad, both for himself and for his fragile 
and insecure social system, which he values and which gives meaning and color to his 
life? And is there any ground for thinking that what is logical and reasonable to you 
will be so to him? If you do not know his axioms, it is probable that your axioms will 
not be his, and, besides, he cannot, in general, be persuaded by reason but only by 
an appeal to his sentiments. How shall you appeal to sentiments that you do not know? 

The stability of the world in which he feels — however non-logically, however 
irrationally — that he is living — however vague, however unanalyzed his feeling — 
is of first importance to every man. That world is largely made up of his human social 
relations and of the sentiments that habitually arise in the course of his habitual 
performance of the routines and rituals of daily life. Change too rapidly imposed 
from without, for him is evil, because his social system cannot change very rapidly 


without breaking; it is bound together by sentiments, which change slowly and resist 
change, because rapid change is destructive of routines and rituals, of habits and of 
conditioned behavior. Such change is painful even to a dog. 

The social environment is what sentiments, routines and rituals make it. From 
the most perfect family in the world take away the sentiments, the routines and the 
rituals, and the residue will be unrelated individuals. No doubt the social environment 
(the various social systems) of a factory is in many ways less important and far less 
perfect than the social system of a good family. But in several respects it is the same 
kind of thing and, as experiment shows, it is in several respects so important that it 
cannot be neglected by anyone who wishes to plan wisely, or even merely to know 
what he is doing. 

The environment is at once physical, chemical, biological, psychological, eco- 
nomic and sociological. As a rule, we all have the strongest feelings about its sociologi- 
cal properties and the least intellectual awareness of them. Often these are the most 
important properties of the environment. Let us study, weigh, modify and use them. 







Our subject tonight is Control. Of course that is what we have been talking about 
all along — when we were considering orders or authority or leadership or co-ordi- 
nation. This final talk therefore may be considered partly as a summary of the other 

In our best managed industries, we notice two points about control: (1) control 
is coming more and more to mean fact-control rather than man-control; (2) central 
control is coming more and more to mean the correlation of many controls rather than 
a superimposed control. 

In regard to the first point, notice how often the word control is used in the sense 
of fact-control. We hear, for instance, of Inventory Control. We used to think the 
Inventory helped us in our control, we did not talk of it as if it were in itself a control. 
We hear, again, of Budgetary Control. This means that where you have cost-account- 
ing and unit-budgeting, the general manager and the head of a department are both 
subject to an impersonal control. The head of a department does not receive an arbi- 
trary order from the general manager, but both study the analyses and interpretations 
which cost- accounting and unit- budgeting have made possible. 

Control is becoming less personal in the old-fashioned sense; control and fact- 
control are becoming synonymous. 

My second point was the correlation of controls. The ramifications of modern 
industry are too wide-spread, its organization too complex, its problems too intricate 
for it to be possible for industry to be managed by commands from the top alone. This 
being so, we find that when central control is spoken of, that does not mean a point 
of radiation, but the gathering of many controls existing throughout the enterprise. 

Genuine control then, that is, fact-control and correlated control, is within the 

With this in mind as our guiding thought, namely, that each situation should 
generate its own control, and in the light of our previous talks, what can we say are 
the principles of control? This is the same as asking what are the principles of organiza- 
tion. For the object of organization is control, or we might say that organization is 

Four fundamental principles of organization are: 

1 . Co-ordination as the reciprocal relating of all the factors in a situation. 

2. Co-ordination by direct contact of the responsible people concerned. 

3. Co-ordination in the early stages. 

4. Co-ordination as a continuing process. 

My first principle, co-ordination as the reciprocal relating of all the factors in a 
situation, shows us just what this process of co-ordination actually is, shows us the 
nature of unity. We considered unity last week, but I want tonight to penetrate further 
into its meaning. We saw last week the process by which any two people may com- 
bine their different kinds of knowledge and experience. I compared it to a game of 



tennis. Let us now take more than two. There usually are more than two concerned 
in any decision. Take four heads of departments. You cannot envisage accurately 
what happens between them by thinking of A as adjusting himself to B and to C and 
to D. A adjusts himself to B and also to a B influenced by C and to a B influenced by 
D and to a B influenced by A himself. Again he adjusts himself to C and also to a C 
influenced by B and to a C influenced by D and to a C influenced by A himself — 
and so on. One could work it out mathematically. This sort of reciprocal relating, 
this interpenetration of every fact by every other fact, and again by every other fact 
as it has been permeated by all, should be the goal of all attempts at co-ordination, 
a goal, of course, never wholly reached. 

You will understand that I am simplifying when I speak of A, B, C and D ad- 
justing themselves to one another. They are of course at the same time adjusting 
themselves to every other factor in the situation. Or it would be more accurate to say 
that all the factors in the situation are going through this process of reciprocal relating. 

If anyone finds this principle difficult to accept, I would suggest that it is a prin- 
ciple which he has already accepted in regard to facts. Any fact gains its significance 
through its relation to all the other facts pertaining to the situation. For instance, if 
you have increased sales, you are not too pleased until you find out whether there 
has been an increased sales cost. If there has been, or one out of proportion to sales, 
your satisfaction disappears. Merchandizing shows you this principle at work. For 
merchandizing is not merely a bringing together of designing, engineering, manufac- 
turing and sales departments, it is these in their total relativity. 

This may seem a rather clumsy phrase, total relativity, but I am trying to express 
a total which shall include all the factors in a situation not as an additional total but 
as a relational total — a total where each part has been permeated by every other 

The possible examples from business management of the working of this funda- 
mental principle are innumerable. Take a situation made by credit conditions, cus- 
tomers' demand, output facilities and workers' attitude. They all together constitute 
a certain situation, but they constitute that situation through their relation to one 
another. They don't form a total situation merely by existing side by side. 

It is necessary to emphasize this because while it is customary nowadays to speak 
of "the total situation" — you find that phrase often in articles on business manage- 
ment — that phrase, total situation, means to many people merely that we must be 
sure to get all the factors into our problem. But that is by no means enough for us to 
do, we have to see these factors each one affecting every one of the others. 

Many examples of this come to mind at once. Take an instance of a social worker. 
She is dealing with a girl of a difficult temperament, who has a nagging stepmother, 
is working at a job for which she is not fitted, and has evening recreations of not the 
most wholesome character. It is obvious that here you have a situation, a whole, 
made up not of its parts but of the interacting of the parts. Perhaps it is because the 
girl is working at something she is not interested in that makes her seek over-excite- 
ment in the evening. And so on. The most successful social worker is not the one who 
deals with these separately, but who sees them in relation to one another. 

This is the first requirement of statesmanship. We shall get no control over eco- 


nomic conditions until we have statesmen who can meet this requirement. We shall 
get no grip on our economic affairs until we acquire a greater capacity than we seem 
to have at present for understanding how economic factors interpenetrate at every 

I am talking in all this of the nature of a unity. This, which is a matter of every- 
day experience to business men in their problems of co-ordinating, happens to be con- 
sidered by some scientists the most important thing in present scientific thinking. 
The most interesting thing in the world to me is the correspondence between progres- 
sive business thinking and certain recent developments in the thinking of scientists 
and philosophers. Such biologists as G. S. Haldane, such philosophers as Whitehead, 
such physiologists as Sherrington, are telling us that the essential nature of a unity is 
discovered not alone by a study of its separate elements, but also by observing how 
these elements interact. 

I could give you many examples from the sciences. I am going to take a moment 
to give you one, although it may seem far from my subject, simply to bring home to 
you this remarkable correspondence in thinking in such entirely different fields. I 
found this in an article in a Journal of Zoology — a very different subject from busi- 
ness management! The article was on the local distribution of wild mice, and the 
whole point of the article was that this distribution, while controlled by food and water 
supply, by nesting material, by climatic conditions and by antagonism between 
species, while controlled by these, was controlled by them only as they were related 
to one another, that the behavior of the wild mice was governed by an environmental 
complex, that it was not influenced by these various factors one by one. 

I thought this expression, "environmental complex," strikingly like what I have 
been trying to say to you in describing the nature of unities, very much like what I 
called a relational total as distinct from an additional total. And business men, as I 
have said, see this every day. The ablest business man, or social worker, or statesman, 
the ablest worker in any field, looks at an "environmental complex," sees the solution 
of his problem depending on the interacting of the elements of that complex. 

This seems to me a principle of the utmost importance for industry or for any 
joint endeavor. This seems to me as important a principle for the social sciences as 
Einstein's theory of relativity has been for the natural sciences. They are both, it may 
be noticed in passing, concerned with relativity. I believe that the principle of rela- 
tivity in the realm of social theory will displace as many of our old ideas in the social 
sciences as Einstein's has in the natural sciences. And I think it greatly to the honor 
of progressive business thinking that it is taking a lead here — a lead which I am 
sure must be followed eventually by statesmen, national and international. 

Before I leave this point, let me call particularly to your attention that this re- 
ciprocal relating, co-ordinating, unifying, is a process which does not require sacrifice 
on the part of the individual. The fallacy that the individual must give up his indi- 
viduality for the sake of the whole is one of the most pervasive, the most insidious, 
fallacies I know. It crops up again and again in one place after another. In some of 
the businesses I have studied, I have been told that the head of a department should 
subordinate the good of his department to the good of the whole undertaking. But of 
course he should do no such thing. His departmental point of view is needed in the 


whole. It must indeed be reconciled with all the other points of view in the business, 
but it must not be abandoned. Just as we have been told by an eminent authority in 
international matters that men should not denationalize themselves but international- 
ize themselves, so I should say to the heads of departments that they should not de- 
departmentalize themselves but inter-departmentalize themselves. In other words, 
departmental policy should be an integral part of what is known as "general policy." 
General policy is not an imaginary "whole," an airplant, it is the interweaving of many 
policies. Whether we are talking of the individual man, or individual department, the 
word should never be sacrifice, it should always be contribution. We want every 
possible contribution to the whole. 

My second principle was co-ordination by direct control of the responsible people 
concerned. We saw last week that in some industrial plants, control is exercised 
through cross relations between heads of departments instead of up and down the 
line through the chief executive. This seems sensible, as these are the people closest 
to the matter in hand. Moreover, if my first principle was right, if the process of co- 
ordination is one of interpenetration, it is obvious that it cannot be imposed by an 
outside body. It is essentially, basically, by its very nature, a process of auto-controlled 
activity. It is the same as with the individual. We know that every individual has 
many warring tendencies inside himself. We know that the effectiveness of an indi- 
vidual, his success in life, depend largely on these various tendencies, impulses, de- 
sires, being adjusted to one another, being made into one harmonious whole. Yet 
no one can issue a fiat by which I am adjusted, I can only be helped to adjust myself. 
It is the same with a group, with a group of executives, for instance. Here, too, the 
process is one of self-adjustment. This being so, it is essential that they should have the 
opportunity for direct contact. 

My third principle was co-ordination in the early stages. This means that the di- 
rect contact must begin in the earliest stages of the process. We see how this works in 
the correlation of policies in a business. If the heads of departments confront each 
other with finished policies, agreement will be found difficult. Of course they then 
begin to play politics, or that is often the tendency. But if these heads meet while 
they are forming their policies, meet and discuss the questions involved, a successful 
correlation is far more likely to be reached. Their thinking has not become crystal- 
lized. They can still modify one another. Their ideas can interweave. I should say that 
one of the fundamental ideas for business management is that the making of decisions 
and the correlation of decisions should be one process. You cannot, with the greatest 
degree of success for your undertaking, make policy forming and policy adjusting two 
separate processes. Policy adjusting cannot begin after the separate policies have been 

I speak of the correlation of departmental policies, yet the principle of early 
stages should, I believe, begin to be operative far earlier than with the heads of de- 
partments — with heads of sub-divisions, with foremen, and, where you have union- 
management, co-operation with the workers themselves. In the union-management 
plan of the Baltimore & Ohio Railroad, the adjustment of trade unions and manage- 
ment begins down in the lowest shop committees. We see this also in the Canadian 
railways. The same principle should guide us where we have shop stewards or em- 


ployee representatives on committees. That is, we shouldn't put to these representa- 
tives of the workers finished plans in order merely to get their consent. We should 
bring them into the game while the plan is still in a formative stage. If we don't, 
one of two things is likely to happen, both bad: either we shall get a rubber-stamped 
consent and thus lose what they might contribute to the problem in question, or 
else we shall find ourselves with a fight on our hands — an open fight or discontent 
seething underneath. 

These two principles — direct control and early stages — which I have seen in 
operation in some of our industries, governed some of the Allied co-operation during 
the War, and are vigorously advocated by Sir Arthur Salter in his "Allied Shipping 
Control." He thinks that adjustments between nations should be made not through 
their Foreign Offices, but between those who exercise responsible authority in the 
matters concerned, that is, between departmental ministers. This corresponds, you 
see, to what I have said of the cross-relations between departments in a business. 
And in regard to the principle of early stages, Sir Arthur shows us most convincingly 
that a genuine international policy cannot be evolved by first formulating your na- 
tional policy and then presenting it as a finished product to confront the policies of 
other nations. For the only process, he tells us, by which completed policies can be 
adjusted is that of bargaining and compromise; if you want integration, he says, the 
process of the interpenetration of policies must begin before they are completed, 
while they are still in the formative stage. 

It seems to me extraordinarily significant that we should find these principles 
recognized in such different fields as those of business management and international 
relations. It means that our ablest thinkers, men who are at the same time thinkers 
and doers, have found a way of making collective control collective ^//-control — 
that is a phrase used by Sir Arthur and I think it a remarkably good one — collective 

My fourth principle was co-ordination as a continuous process. Just as I think 
that co-ordination cannot be enforced on us, but must be done by ourselves, just as 
I think it must begin in the earliest stages, so I think it must go on all the time. I 
do not think that the various people concerned should meet to try to unite only when 
difficulties arise. I think that continuous machinery for this purpose should be pro- 

One reason for this is that there is then a greater incentive to discover the prin- 
ciples which can serve as guides for future similar cases. If an industrial plant has 
continuous machinery for co-ordination, and if it makes some classification of prob- 
lems, then when a fresh problem arises it will be able to see the points in which that 
resembles a certain class of problems and it can ask, "Have we evolved any principles 
for dealing with problems of that kind?" One of the interesting things about the 
League of Nations as one watches its work at Geneva is that many in the Secretariat 
are trying deliberately to discover the principles underlying the decisions made in 
order that they may be taken as precedents in similar cases arising later. A member 
of the political section of the Secretariat said to me: "Our treatment of every question 
is two-fold: (1) an attempt to solve the immediate problem; (2) the attempt to dis- 
cover root causes to help our work in the future." 


Another advantage of continuous machinery for co-ordination is that then the 
line is not broken from planning to activity and from activity to further planning. A mistake 
we often tend to make is that the world stands still while we are going through the 
process of a given adjustment. And it doesn't. Facts change, we must keep up with 
the facts; keeping up with the facts changes the facts. In other words, the process of 
adjustment changes the things to be adjusted. If you want an illustration of this, con- 
sider the financial and economic adjustments between nations. When one financial 
adjustment is made, that means only that we have a fresh financial problem on our 
hands, the adjustment has made a new situation which means a new problem. We 
pass from problem to problem. It is a fallacy to think that we can solve problems — in 
any final sense. The belief that we can is a drag upon our thinking. What we need is 
some process for meeting problems. When we think we have solved a problem, well, by 
the very process of solving, new elements or forces come into the situation and you have 
a new problem on your hands to be solved. When this happens men are often discour- 
aged. I wonder why; it is our strength and our hope. We don't want any system that 
holds us enmeshed within itself. 

In order, however, to get the fullest benefit of continuous machinery for co-ordi- 
nating, in order to utilize our experience, get the advantage of precedents, be able to 
formulate principles, we must learn how to classify our experience. I do not think any 
satisfactory method for that has yet been worked out. I was present once at a meeting 
of heads of departments in a large shop, and heard one of these heads say in regard to 
a case they were discussing, "We had a problem like this two or three years ago. Does 
anyone remember how we treated that?" No one did! We talk much about learning 
from experience, but we cannot do that unless we (1) observe our experience, (2) keep 
records of our experience, and (3) organize our experience, that is, relate one bit to 
another. Unrelated experience is of little use to us; we can never make wise decisions 
from isolated bits, only as we see the parts in relation to one another. 

I have given four principles of organization. These principles show the basis of 
control, show the process of control, show that control is a process. They show us con- 
trol as self-generated through a process of the interweaving of the parts. The degree of 
correlation is the measure of control: the more complete the reciprocal adjusting, the 
more complete the control. 

We find this principle also in the sciences. I said a few moments ago that scientists 
are finding that the nature of the unities they deal with is governed by a certain prin- 
ciple and that that is the same principle which we find in the co-ordinations that ap- 
pear in the running of a business. I want now to go further and say that both scientists 
and business men find that this principle of unity is the principle of control, that is, 
that organization is control. 

Biologists tell us that the organizing activity of the organism is the directing 
activity, that the organism gets its power of self-direction through being an organism, 
that is, through the functional relating of its parts. 

On the physiological level, control means co-ordination. I can't get up in the 
morning, I can't walk downstairs without that co-ordination of muscles which is 
control. The athlete has more of that co-ordination than I have and therefore has more 
control than I have. 


On the personal level I gain more and more control over myself as I co-ordinate 
my various tendencies. 

This is just what we have found in business. Let me remind you how often we 
have noticed this even in the few illustrations I have had time to give you in these 
talks. We saw last week that if the price of a certain article has to be lowered, the 
situation will not be controlled by the production manager's solution of the problem, 
nor by the sales manager's. The situation will be controlled when these two men and 
the others concerned unite their different points of view. We saw that if the personnel 
manager tries to force his opinion of a worker on the foreman, or the foreman tries to 
force his point of view on the personnel manager, the situation will not be controlled. 

The question of the debt to America will find no satisfactory solution if either 
America tries to force her will on England or England tries to force her will on 
America. We shall have control of the situation only if England and America are able 
to unite their different points of view, only if we can find what I have called in these 
talks an integration. A writer in the Observer last Sunday spoke of the divergence be- 
tween English and American opinion on the debt question, but added that there were 
indications that we might yet be able to get the hyphen back into Anglo-American 
opinion. That expresses wittily and concisely what I have taken two hours to say to 
you. His hyphen is a symbol of my integration. If instead of an English opinion and an 
American opinion, we can get an Anglo-American opinion, that unity will mean con- 
trol of the situation. 

But this is in the nature of an aside. What I was doing was looking back over our 
four talks to see what indications we had that this conception of unity and control as 
synonymous is gaining ground today in the running of business or of any enterprise. 
One very clear indication, I think, is the fact that in seeking for leaders we are not 
thinking first of the ability to dominate people. The head of one of the women's 
organizations in the War told me that on one occasion when a woman offered herself 
for work, she was asked by the board which received applications what she considered 
her qualifications for an officer's position. She replied, "Because I have had so much 
to do with blacks." The board were amused; they had in their minds quite other 
qualifications for officers' positions than that of being a disciplinarian. 

The leader has not only to deal with people but with situations. I think I may 
say now what I could not say so definitely in my talk on leadership because we had 
not then considered this subject of correlation, that an understanding of the process 
of correlation is one of the chief requirements of leadership. We all see this every day. 
If my maids don't get on well together, it isn't enough for me to command them to do 
so, and it certainly isn't enough for me to preach brotherly love. I must find out just 
what adjustments to each other they find difficult to make and show them how to 
correlate those particular differences. I think this matter of brotherly love has been 
terribly exaggerated, as between people or between nations. There is no use in preach- 
ing that to us unless we are told how to live in brotherly love. I know plenty of people 
who want to, who simply don't know how to. It is of course a good thing for men and 
women of different nations to feel kindly toward one another, but it won't meet 
emergencies unless we know how to integrate our differences, or, to use the language 
of the Observer correspondent, unless we know how to put the hyphen in. 


I had two chief points to make about control. One, which I have been consider- 
ing, that it comes from unity, organization, integration. The second is that control is 
a process. We have had indications of that, too, throughout these talks. Let me men- 
tion one. In my first talk I spoke to you of obedience. It was difficult for me to say 
all that I wished to about that until we had got to this place in our thinking. I think 
we should consider obedience as one moment in a continuous process. If I obey and 
do my work in the best way possible, obey intelligently, that obedience may get 
incorporated in the next order. With this conception you don't see order and obedi- 
ence as two entirely different things. You see them as part of a whole. You see a proc- 
ess; one moment in that process we may call an order, another moment we may call 
obedience. Life is continuous and it is for us to see the connection between one mo- 
ment and another. Life is whole and it is for us to see it whole. To realize that our aim 
is control, and to recognize that obedience contributes to that control as well as the 
order, is I think to get the whole fun out of working together, whether we are the ones 
who obey or the ones who order. 

If you look at business not theoretically, but as it is, you don't find the board of 
directors controlling the general manager and the general manager the sales manager 
and the sales manager the salesman. You see that all the time many are sharing in the 
control, that they are taking part in a process. 

I must take the precaution here to repeat to you what I have said before, that 
as I am talking largely of tendencies in these talks, I can say few things that are a 
hundred per cent true. We know that a board of directors may have a certain policy, 
as, for instance, in regard to labor, and that policy must be accepted throughout the 
plant. But in the daily routine of buying and manufacturing, advertising and selling, 
what I have said is true. 

And if control is the process of the inter-functioning of the parts, if the most 
perfect control is where we have the inter-functioning of all the parts, then I think the 
workers should have a share, not from any vague idea of democracy, not because of 
their "rights," but simply because if you leave out one element in a situation you will 
have just that much less control. It has been found that piece-rates cannot be wholly 
decided by an expert, the question of fatigue cannot be wholly decided by psycholo- 
gists, the cost of living cannot be wholly decided by statisticians. And so on. 

And it is because of this conception of control which I have been giving you that 
I cannot believe in "workers' control" as advocated by some in the Labor Party. 
I think managers and workers should share in a joint control. This to my mind is so 
important that I am sorry it could not have a larger place in these talks. 

We aim then at co-ordination in business because we know that through unity 
an enterprise generates its own driving force. And this self-generated control does not 
coerce. But I do not think that this kind of control is sufficiently understood. Everyone 
knows that our period of laissez-faire is over, but socialists wish to give us in its place 
state control, and they mean by that state coercion — we find again and again in 
their pamphlets the words force, coerce. Those using these words are making the 
fatal mistake, I believe, of thinking that the opposite of laissez-faire is coercion. And 
it is not. The opposite of laissez-faire is co-ordination. 

Others who do not believe in state control are urging National Planning Boards 


of experts to co-ordinate industry. If these boards were to be composed of the heads 
of industry or their representatives we might hope to have the kind of self-adjusting, 
of self-correlating which I have been describing to you, but I have not seen any plan 
which allows for this process. Therefore I do not believe that as at present conceived 
they will bring us any appreciable degree of co-ordination. The policies of our differ- 
ent industrial and economic organizations will have to be adjusted to one another 
by changes not imposed by an outside authority, but voluntarily undertaken, no, not 
exactly undertaken, but spontaneously brought about by the process of interpenetra- 
tion. In order for this to be done, the planning boards will have to be composed of the 
heads of the industries themselves, of course with expert economists on the board as 
well. I think the consideration of planning boards a splendid step in the right direc- 
tion. I am only hoping that before we establish such boards we shall see that both their 
composition and their functions are in line with the more progressive thinking on 

The period of laissez-faire is indeed over, but I do not think we want to put in 
its place a forcibly controlled society whether it is controlled by the State of the social- 
ists or the experts of a planning board. The aim and the process of the organization 
of government, of industry, of international relations, should be, I think, a control 
not imposed from without the regular functioning of society, but one which is a co- 
ordinating of all those functions, that is, a collective self-control. 

If then you accept my definition of control as a self-generating process, as the in- 
terweaving experience of all those who are performing a functional part of the ac- 
tivity under consideration, does not that constitute an imperative? Are we not every 
one of us bound to take some part consciously in this process? Today we are slaves to 
the chaos in which we are living. To get our affairs in hand, to feel a grip on them, to 
become free, we must learn, and practice, I am sure, the methods of collective control. 
To this task we can all devote ourselves. At the same time that we are selling goods or 
making goods, or whatever we are doing, we can be working in harmony with this 
fundamental law of life. We can be assured that by this method, control is in our 

I heard an address by the managing director of a certain firm who said in the 
course of his address that the emphasis in regard to facts used to be on the accuracy 
with which they were gathered, and the fairness and balanced judgment with which 
they were interpreted. Now, he said, we are coming to know that we can make facts. 
It seems to me that there is much food for thought in that sentence. We need not 
wait on events, we can create events. 

I cannot do better than end with some words written by Wells long ago in the 
first chapter of "The New Machiavelli." "It is," he said, "the old appeal indeed for 
the unification of human effort and the ending of confusion. . . . The last written 
dedication of all those I burnt last night was to no single man, but to the socially 
constructive passion — in any man." 











It is comforting to reflect that one of the vastest pieces of industrial organization 
in the history of the world was associated with the beginnings of an undying literature. 
The building of Solomon's temple called for a closeness of organization which amazes 
one the more it is examined. Solomon himself was the chief organizer, with a passion 
for detail. His association with Hiram of Tyre for the production of all sorts of ma- 
terial, gold and silver and cedar and purple, and for the purchase of the services of 
skilled workmen, including Huram-abi, the architect, was the beginning of a func- 
tional division, for there were the skilled technicians, and side by side with them were 
the workers — over 70,000 of the "strangers in the land" to bear burdens, and 80,000 
to hew the stone in the mountains, and 3,600 overseers, probably Israelites, as de- 
serving a more honorable position. One would like to know more about this early 
venture in industrial organization and by what means some of the tremendous stones 
were carved and conveyed, and how the great brass laver resting on twelve oxen, 
with the bases on which they were wheeled from place to place, was brought to 
Jerusalem. Still the broad divisions show us the separation between artistic designing, 
artistic production, the labor of transport and the labor of building with their parallel 
distinctions in supervision. Functional divisions, at least in their essential form, have 
a long history. There is an admirable account in Mr. H. Stuart Jones' article on 
"Administration" in "The Legacy of Rome" of the organization which "without 
ceasing to be the Imperial household, became the Whitehall of Ancient Rome." 
There is the accountant, the principal private secretary, the clerk of petitions, with 
five differentiations of function, and later on we find Hadrian breaking with "the 
idea that the citizen must be equally qualified to render service in peace and war" 
and establishing a purely civil service with functional divisions. So, also, we find 
functional divisions in the ecclesiastical orders, and throughout history we see recog- 
nitions of the fact that organization always demands some functional differentiation. 

As functional divisions are urged upon us today they come from that conception 
of Scientific Management which believed that organization should be focussed upon 
functional divisions. So it was that in Taylor's scheme there were to be eight foremen 
each with his own function. This extreme doctrine has largely been surrendered, and 
functional division takes a part which I think can best be called a co-ordinated part 
in organization. What exactly that co-ordinated part is to be has not been thought 
out. It seems to me to be quite fair to say that functional division of higher responsi- 
bilities has its kinship with that functional division of the work of production which 
we call "Mass production." It comes from an emphasis upon Men in the bulk rather 
than upon Men as individuals. In a sense it is a recognition of human limitation, but 
I would prefer to describe it as a correlation of human capacities into an organic 
whole. I take the following definition from a recent valuable book on Factory Or- 
ganization: "By this plan, specific functions common to all or several departments 
. . . are each placed in the hands of a man specifically qualified for his particular 
function, and instead of giving attention to all the factors in one department, he gives 



his attention to one factor in all departments." It means, obviously, that what ordi- 
narily we have called "departments" are broken up, and there are fresh divisions on a 
functional basis culminating in the management. From the same admirable book we 
may sum up the advantages and disadvantages of functionalization before we come 
to our own analysis. There is no doubt a focussing of an expert efficiency which tends to 
raise the whole tone from the point of view of considered expertness. There is a basis 
for expansion which is not provided by the departmental system without vast change 
and the addition of departments. There is the presumption that men are better suited 
to the function to which they are allotted and that they will be for that reason more 
happy and confident in their work. Then there is the fact that it is a type of specializa- 
tion which is in accord with the spirit of the age. So far for advantages. On the other 
hand, a precise delimitation of function is exceedingly difficult to fit in with an or- 
ganization, and it does happen that a functional organization is very complex. It 
demands a sense of organization which conflicts with many natural human instincts 
and especially with the instinct of control, for to fit in a functional organization de- 
mands self-surrender. There is also a danger lest responsibility should be passed from 
function to function, since it is exceedingly difficult to divide and to define the func- 
tions with sufficient precision. Lastly there is the fear lest the functional division should 
not fit in with human qualities, so that capacities and inclinations may either overlap 
or find themselves lost between functions. 

When we come to examine the position in our own way it is as well to remember 
that functional divisions of authority and of responsibility differ in degree rather than 
in kind. Put as we have put it above there would seem to be a fundamental differ- 
ence between the departmental and the functional system, the former being some- 
times called "geographical" or "territorial." It is very doubtful if there is today a 
purely "geographical" system of division in any industry. At any rate the accounting 
will be separate and will function for all the departments; so will the technical proc- 
esses and at the other end the sales. Then it should be remembered that in life there is 
a considerable adoption of the functional system. The professions are functional and 
indeed as time goes on the functional side is emphasized, as can be seen in the medical 
profession, where it is now customary for two or three medical men to be associated 
together in serving a district, a remarkable change from the "geographical" or "terri- 
torial" to the "functional," and very considerably aided by the telephone and the 
motor-car. We may say, therefore, that in industry we are faced not so much by the 
crude issue between the functional and the territorial systems as by the more or less 
of the functional. Moreover, it will be as well for us to have in mind the danger which 
accrues from striving to build up a final organization at one stroke. "We shall be 
fortunate," says Mr. Stuart Jones, "if the builders of the new order bring to it the 
tact and patience of Augustus, and his infinite capacity for taking pains in framing 
provisional institutions so as to provide for orderly development." When we place a 
perfected departmental system side by side for purposes of comparison with a per- 
fected functional system we are failing to recognize the fact that growth and develop- 
ment are of the essence of any human organization. The error which Taylor made 
in his eight functional foremen was to overlook the human need on the part of the 
workers for such definite direction as is incarnated in one person. It is true that to 


have a foreman who in turn appeals to eight functional chiefs is to make it but one 
remove; it is, however, just that one remove which makes all the difference. It was 
the intrusion of some of the departmental element which made the functional or- 
ganization possible. 

On this ground, therefore, I would hesitate to put into crude and rival divisions 
the pros and cons of the functional system. It will be wiser, I think, to take it for granted 
that the functional system is inevitable, just as mass-and-machine production was 
inevitable. But it is the work of human beings to safeguard humanity against many 
ills which seem to accompany movements apparently inevitable. A full-fledged 
functional system, vigorously introduced, is full of peril. In the first place it tends to 
dethrone the discipline of industry upon the human mind. It operates in this direction 
by its tendency to rob the industrial worker and his immediate director of the sense 
of a completed product resulting from their work. Put a little differently, it has in 
respect of direction precisely the demerit of mass production in workmanship. In 
the second place, though it would seem, on the face of it, to allot responsibilities ac- 
cording to particular capacities, it does leave gaps, and the more precise the functional 
division, the more likelihood there is of these gaps and of friction in respect of opera- 
tions with regard to which it is doubtful whether they belong to this or to that func- 
tion, or belong in a measure to more functions than one. This possibility of friction 
does much to counterbalance the increase of skill and of zeal and of happiness which 
follow from the allotment of responsibilities according to personal qualifications. In 
the third place there is a lack of elasticity; it follows in many cases that an increase 
of knowledge on the part of a functional director tends to bring him up against the 
hard walls which limit his functions. Lastly, there can be little doubt that to the really 
able functional director a long experience tends to irritation and to discontent; and 
discontent on the plane of leadership is a very grave evil. It is not always the case 
that the function fills all that he finds himself able to do. He peers over the walls, as 
into a Paradise, and he regards his reputation in one particular line as being almost 
a disaster to him. If only, he sighs, he could change his function for a while — but 
then it is of the very essence of the functional system that the functions are permanent 
or, at any rate, that they are not readily changeable. 

So that what we stand for in constructive criticism of the functional system is to 
seek such an application of the system as will be least disadvantageous to human-kind. 
Manifestly, the continuous intensive development of industry and the increasing com- 
petition force upon us the adoption of every means to efficiency, and among those 
means the functional system of direction is inevitable. Where this system has been 
introduced to the best purpose the greatest care has been taken to procure co-ordina- 
tion, sometimes by committees and sometimes by special co-ordinating functional 
directors. In its origin, however, as laid down by F. W. Taylor it was a much simpler 
matter than it has become, for it was "a recognition of qualitative differences, as to 
capacities required and the conditions of routine performance, between the function 
of execution in manipulating machine or tool and the function of planning and pre- 
paring work; and the grouping of operatives into two major divisions responsible 
respectively for these qualitatively different groups of functions." Thus the present 
claims for somewhat elaborate differentiation of function have grown from the funda- 


mental differentiation into two, where probably all would be in agreement and where 
co-ordination was not so difficult. It is co-ordination which is the crux of the problem 
today. We may smile at the suggestion that a specially expert staff has been necessary 
in some cases to co-ordinate the precedent work of experts. It is clear, however, that 
functional development is only practicable, or at any rate is more likely to be success- 
ful, when it is firmly based upon a departmental division where co-ordination is secured 
to begin with. On this subject of organization we have much to learn, in general, 
from the Army, but in no particular is the lesson more valuable than in the means 
of weaving a functional system into a departmental system. Signals, communications, 
ordnance, aeroplanes, tanks — all of them highly technical modes of warfare — have 
successfully been brought into play, and they have their proper subordination to the 
general scheme while they have their full functional development. We may apply 
this lesson to industry in a spirit of paradox by saying that we shall get the best results 
from an adoption of the functional method where we do not attempt to substitute it 
at once and in toto for the departmental or geographical or territorial method. Prob- 
ably much of the criticism which the functional principle has received has been due to 
this preliminary misunderstanding. It has been regarded as an alternative method of 
organization from top to bottom, and men trained by long years of experience and of 
habit to the other systems have found it difficult to adapt themselves to it. We should 
have made more progress if we had proceeded a little more slowly. We had forgotten 
that for the most part the organization under the old conditions was largely haphaz- 
ard, and that we needed to teach the value of organization itself before introducing a 
principle which depends upon the sense of organization. The balance between func- 
tional direction and general direction has to be found in the particular industry, and 
it has to be found at various levels, and it is in the effort to discover this balance that 
we shall learn to develop the sense of organization and to find the true place for a 
functional system. 

Then there is the question of the individual. Men and women who are conscious 
of their ability, who have the sense of responsible leadership, are bound to feel that 
the functional system robs them of scope. They have been accustomed to looking at 
the industry as a whole, at its processes as a whole, at its production as a whole. They 
find themselves not only focussed in their attention but actually robbed of other points 
of view. They feel that though all along they may have had a specialized functional 
interest, yet that it was accompanied by healthy marginal interests, and that these 
are ruthlessly cut off. Instinctively they regard themselves as sacrificed, under a func- 
tional system, to the success of that system. To these criticisms, weighty as they are, 
we have to make the reply that the functional system has some place in the adaptability 
of industry to be the chief force which is fashioning us in our time. Whether it be a 
tradition from the Individualism of last century or not, it is probable that our lack of 
social cohesion today is largely due to the fact that in our industrial enterprises we 
are not yet trained in social cohesion. It may be that the functional organization in 
future will play its part in welding us together more closely and more mutually. 
Plato told us, as St. Paul told us, that "there are diversities of natures amongst us 
which are adapted to different occupations," that "all things are produced more 
plentifully and easily and of a better quality when one man does one thing which is 


natural to him and is done at the right time and leaves other things." Possibly that 
referred rather to professionalism of calling than to functionalization as we understand 
it in the unit industry. Yet there is reason to suggest that just as this moderate function- 
alism was the basis of Plato's claim for a co-ordinated State, so the interweaving of 
functions and the close inter-dependence of human unit with human unit may come 
about from an acceptance of the functional system in industry in such a way that it 
may develop, but also that it may be co-ordinated. "To the Greek mind the city or 
state, convertible terms, was a community — an association of men — an ethical 
unity. It was this concept which dominated the social theory of Plato and Aristotle. 
Civic life was the normal life, and it was the relations which men sustained to each 
other and to the collective whole that constituted the problems of social philosophy." 

They constitute also the fundamental problem of industrial administration. I 
have described functional organization as inevitable. I believe that it is inevitable 
if there is to be organized efficient production. But there is reason, at the same time, 
to be apprehensive if it is introduced as a completed whole without regard to its 
corollaries. So essential is co-ordination that I would plead for the continuance of 
the departmental organization as the basis and the introduction of the functional 
system, gradually, and only so far as the "sense of organization," to use words which 
Dr. Northcott uses in the book I have already quoted, justifies the venture. Nor is it 
certain that there are types of mind which suit the functional and others which suit 
the departmental. It may indeed prove to be wise, when we know something more 
about functional systems in practice, so to organize the functional tasks that men may 
pass from them to departmental tasks, or may pass from one functional task to an- 
other after a reasonable, probably rather a long, period. There may be an apparent 
loss at the time, but the widening and the freshening of outlook may prove to be an 
ample recompense. For it is not only that we are considering the influence of our 
method of industrial organization upon the men and women to whom we give the 
exercise of authority, but that we are remembering that this, in turn, will re-act and 
that the industry of tomorrow will gain. The precision of the detail which will be 
within the scope of the knowledge of the functional practitioner will be all the health- 
ier when it is balanced by being transferred to the point of view of general manage- 
ment, or when, perhaps, it is correlated to the detail of the knowledge of another func- 
tional position. We have leaped rather too readily to the view that the functional 
positions are finally and permanently separated from each other and from the manage- 
ment. We have leaped far too hurriedly to the conclusion that the functional divi- 
sions, as we have made them to suit our industry, correspond to some basic psycho- 
logical differences. It may be that the differences are not basic, and that a higher 
culture for industry is practicable by means of a functional system which has been 
weaved into a departmental organization. It may be that by some such means as 
these we can aid in the necessary task of co-ordination, and that we can use men, far 
more highly trained, and with actual experience in a few functional responsibilities, 
in the work of general direction, to the advantage of that direction. 

Thus I am not at all ready to take up the cudgels either for or against the func- 
tional system. In the rush of enthusiasm in which it was introduced there were very 
real dangers, and these dangers were so evident that many were led rather hastily 


to scrap functionalization altogether. But it is clear that in some way or other we 
must provide for a close specialization and that where the same type of specialization 
is needed in various departments of an industry a functional allotment is inevitable. 
Not only is it specialization but it is particularization of a point of view. Wise man- 
agement can see this coming, and set out to meet it without making a sudden change 
from departmental or territorial organization to purely functional organization. To 
go out and meet it is one thing, to provide for it, to test it within the range of its adop- 
tion so that its value for a wider range will be known — this seems to indicate the 
sound process. But it will depend upon the industry and it will depend upon the man- 
agement. For we shall do well to ask ourselves how it comes about that in our in- 
dustries, with all the opportunities which we have had for closer organization, in the 
main we have failed to procure any similitude of social cohesion. For this purpose 
there can be little doubt that functional organization is an effective means, since it 
demands personal skill and the frank recognition of the skill of others. In this way 
industry may yet take its place in the training of mankind. 

Though I claim that functional direction is inevitable, just as more and more 
finely distributed mass production is inevitable, I do not think that, standing alone, 
these will be tolerable as the method of industry. The instinct of man will seek some- 
thing which makes a wider demand upon his capacities. So if we are convinced that 
functional direction and machine-mass-production are essential to the successful 
conduct of industry and to the supply of an ever-increasing demand, we shall be 
bound to accompany both the direction and the work of production by other methods 
into which mankind can throw capacities and energies of a different kind. It is at 
this point that we can reach a summary judgment. Functionalization is an essential 
instrument in the conduct of industry, an essential element in organization. But it is 
an instrument in the hands of man, and an essential element in an organization which 
is to be subordinate to man. By what means mankind will be able to use functional 
direction in industry and yet keep himself free from its evils, if it should obtain the 
dominance, is not easy to say at this juncture. We have not yet gone far enough from 
the departmental or the territorial method to be able to conjecture what would be 
needed if functional organization became dominant; precisely in the same way we 
have not yet gone far enough in respect of mass or machine production to be able to 
estimate its appropriate limits. Nevertheless, we are able to peer into the future, and 
we can make something of a guess as to the direction in which correctives for func- 
tional organization, as for mass production, will be found. Clearly they will be of the 
nature of corporate efforts at an understanding of the full purpose of the industry; 
clearly, too, they will use some machinery of the corporate type for the focussing of a 
common attention upon the central purpose, accompanying it by manifold contribu- 
tions from experience and knowledge. If it means therefore that functional organiza- 
tion makes this demand upon us, if it quickens our interest and warms our human 
associations, it may be that it will achieve far more than its votaries dared dream. 
It may lead the way to cheaper and better-organized production and thus to further 
triumphs of industry, but at the same time it may be the means of a revolt against 
the very narrowness of life, both in direction and in production, for which it may seem 
to have been responsible. Those who introduce functional direction with some such 


vision in their minds may be building far more securely than they know, and the 
structure after it is reared may be a true and real association, not merely in produc- 
tion but in that mutuality of endeavor which brings the best out of men and by that 
means builds for the generations which are to come. We know something of the 
legacy of Greece and of Rome and of the Middle Ages. It is clear that the legacy 
which we shall leave behind us is the legacy of industrial organization, far more in- 
clusive in its scope than we yet realize. If we are to hand it on with due and appropri- 
ate and balanced and sensitive regard for the men and women who are to use it as a 
mighty instrument, we shall need to consider in what way it shall be constructed, 
not only to permit and to encourage each individual to do his best, but to permit 
and encourage the corporate whole to be more effective, by reason of the greater 
efficiency of the individual as individual and as part of the corporate whole. This is 
the task before us and we must watch all our experiments from this angle. It subtends 
more of human life than we suppose. It means that man is learning from industry in 
what manner he shall live, and that involves a recognition both of the corporate and 
of the individual aspects and of a tertium quid, the sacredness of the individual aspect 
as part of the corporate aspect. 

A very effective parallel may be drawn between colonization and industry to 
illustrate this point. The task of British colonization today is infinitely more difficult 
and infinitely more complex than the colonization of ancient Rome. It is not because 
the distances which separate the Mother country from the colonies are immeasurably 
greater, but because the establishment of Britain's overseas must involve a nationhood 
of their own which yet shares in the nationhood of the home country and of the whole. 
Rome colonized the world and held her colonies in a firm grip. Today there must be 
something more than a strong grip from the center; there must be a radiant life puls- 
ing through the whole. In industry the old departmental divisions will have many of 
the characteristics of the Roman colonies, and in so far as we use the functional system 
wisely it will seek the channels for a flow of life both outwards and inwards of which 
the purely departmental system did not prove itself capable. Yet in doing so we must 
remember that the functional system itself is an instrument only, and that it is not life 
but only a channel of life. This sums up the position which we shall claim for it. As 
yet no one can define the limits, but we can define the purpose. It may aid us to a 
far closer welding of the industrial structure, enabling all men's contributions more 
readily to be offered and more readily to be accepted. With this as our criterion we 
may well proceed, knowing that, as yet, we are in a very rudimentary stage in respect 
of knowing and understanding of what it is capable. But with an appropriate sense 
of balance, with a recognition of the sense of organization in which it can take its 
place, we can boldly adopt this and that experiment, provided that the sacredness 
of man is kept steadily before us, and that his culture and his development are defi- 
nitely regarded as part of industrial culture and industrial development. Where 
functionalization unites the two it is a helpful ally; where it separates them it is an evil. 




Management Consultant, Paris 






It has long been known empirically to students of organization that one of the 
surest sources of delay and confusion is to allow any superior to be directly responsible 
for the control of too many subordinates. Armies have observed this principle for 
centuries. "The average human brain finds its effective scope in handling from three 
to six other brains. If a man divides the whole of his work into two branches and dele- 
gates his responsibility, freely and properly, to two experienced heads of branches he 
will not have enough to do. The occasions when they would have to refer to him would 
be too few to keep him fully occupied. If he delegates to three heads he will be kept 
fairly busy, whilst six heads of branches will give most bosses a ten-hour day. . . . 
Of all the ways of waste there is none so vicious as that of your clever politician trying 
to run a business concern without having any notion of self-organization. One of 
them who took over Munitions for a time had so little idea of organizing his own 
energy that he nearly died of overwork through holding up the work of others; i.e., by 
delegating responsibility coupled with direct access to himself to seventeen subchiefs. 
... As to whether the groups are three, four, five or six it is useful to bear in mind 
a by-law; the smaller the responsibility of the group member, the larger may be the 
number of the group — and vice versa. . . . The nearer we approach the supreme 
head of the whole organization, the more we ought to work towards groups of three; 
the closer we get to the foot of the whole organization, the more we work towards 
groups of six." x 

It is less well-known in business, though the subject has been mentioned occa- 
sionally in management literature. "At a dinner the other evening I heard the Presi- 
dent of the General Electric Company asked how many people should report directly 
to the President of a large industrial company. He said that eight or nine were re- 
porting at present, but that it was too many, and he was reorganizing his functions 
so that only four or five would report direcdy to himself; and I imagine that four or 
five is enough. Not that a chief executive should not have contact with others; but 
that is about as many general functions as should regularly and directly lead up to 
him." 2 But as this quotation indicates, the question is regarded as an open one. 
The principle is not accepted as final, a definite rule which should be followed by all 
those who seek to administer economically and effectively. 

As long as this is so, wastes in organization arising from the multiplication of 
direct subordinates are likely to continue. Access to the highest possible superior 
offers opportunities for advancement and is itself in many cases a public acknowledg- 
ment of status. Personal ambition, therefore, exercises a constant pressure towards 
indefinite arrangements and away from clear-cut structure. Superiors who are them- 
selves anxious to enhance their prestige and influence in an organization can do so 
most readily by increasing their area of control, adding sections and departments to 
their responsibilities with the least possible emphasis on the organization anomalies 

1 Sir Ian Hamilton, "The Soul and Body of an Army." Arnold, London, 1921, p. 229. 

2 H. P. Kendall, "The Problem of the Chief Executive," Bulletin of the Taylor Society, Vol. 7, No. 2, April, 1922. 



thereby created. Strong personalities are sometimes unready to delegate and endeavor 
to exercise a direct personal supervision over far too many details. Considerations of 
a political character suggest that every kind of "interest" should be represented in 
contact with superior authority. Attempts to simplify situations where a superior has 
too many subordinates are almost invariably regarded as an attack upon his personal 

It is, therefore, of great importance for the art and science of organization that 
evidence of the validity of the principle as a matter of theory, should be added to the 
practical experience of those specially interested in organization. In fact, such theoret- 
ical evidence is overwhelming. It rests upon two simple considerations. The first is 
what is known by psychologists as "the span of attention." Generally speaking, in any 
department of activity the number of separate items to which the human brain can 
pay attention at the same time is strictly limited. In very exceptional cases, for in- 
stance, an individual can memorize groups of figures of more than six digits when read 
out and can repeat them accurately after a brief interval. But in the vast majority of 
cases the "span of attention" is limited to six digits. The same holds good of other 
intellectual activities. 

It is, however, the second consideration which has caused the greatest confusion 
on this question. In almost every case the supervisor measures the burden of his re- 
sponsibility by the number of direct single relationships between himself and those 
he supervises. But in addition there are direct group relationships and cross relation- 
ships. Thus, if Tom supervises two persons, Dick and Harry, he can speak to each 
of them individually or he can speak to them as a pair. The behavior of Dick in the 
presence of Harry or of Harry in the presence of Dick will vary from their behavior 
when with Tom alone. Further, what Dick thinks of Harry and what Harry thinks of 
Dick constitute two cross relationships which Tom must keep in mind in arranging 
any work over which they must collaborate in his absence. The presence of these 
relationships other than the single direct relationship is not always obvious. Yet the 
popular expression "he's no good in a crowd" refers to just such changes in human 
personality as the result of association. And they must constantly, and often simul- 
taneously, constitute additional factors to be controlled. 

Thus, even in this extremely simple unit of organization, Tom must hold four to 
six relationships within his span of attention: 

Direct Single Relationships 

Tom to Dick and Tom to Harry 2 

Direct Group Relationships 

Tom to Dick with Harry and Tom to Harry with Dick 2 

Cross Relationships 

Harry with Dick and Dick with Harry 2 

Total Relationships 6 

The number of direct and cross relationships will increase in mathematical re- 
lation as the number of subordinates assigned to Tom increases. The direct single 
relationships increase in the same proportion as the number of subordinates assigned. 
Each person added creates only one single direct relationship. Direct group relation- 


ships can be counted either once for each possible combination of subordinates, or 
once for each individual in each possible combination. Similarly cross relationships 
can be counted once as "unilateral" or twice as "bilateral." In any event the group 
and cross relationships increase more rapidly than the number of subordinates as- 
signed, because each fresh individual adds as many more cross and direct group 
relationships as there are persons already in the group. Irrespective of the manner of 
counting, the number of relationships increases in exponential proportion. 

The effect of these distinctions as brought out in the accompanying tables and 
chart, should be read as follows: 

n — number of persons supervised; 

a — number of direct single relationships; 

b — number of cross relationships; 

c — ■ number of direct group relationships: 

d — a-\-b; 

e — a-\-c; 


computed on the maximum basis as indicated above. In the second table b\ c\ d\ 
e\ and/', indicate similar figures computed on the minimum basis. 

Since it is not possible to assign comparable weights to these different varieties 
of relationship, it is probably safest to accept the most inclusive assumption as the 
standard by which to judge the relative complexity of supervision imposed by varying 
numbers of subordinates. This is represented by line/ on the chart. Assuming that it 
is possible for one supervisor to watch a maximum of 12 cross and 28 direct group 
relationships, the conclusion follows that, in cases other than routine work, the rapid 
increase of cross and direct group relationships is the governing factor which actually 
limits the number of persons which can be effectually and efficiently supervised by 
one person. Hence the number of lateral divisions in each descending level of responsi- 
bility should be restricted to a maximum of five and, most probably, only four. 

It is of interest also to note the possible number of different relationships in which 
the supervisor can stand to any immediate subordinate. On assumption/, for four 
subordinates there are eleven relationships with any individual, one direct single, 
three cross, and seven direct group. In a group of twelve, 2,059 per member. This 
single fact explains many notorious military disasters. Just why an executive already 
having four subordinates should hesitate before adding a fifth member to the group 
which he controls directly, becomes clear if it is realized that the addition not only 
brings twenty new relationships with him, but adds nine more relationships to each 
of his colleagues. The total is raised from 44 to 100 possible relationships for the unit, 
an increase in complexity of 127 per cent in return for a 20 per cent increase in working 

The exception made in a preceding paragraph for cases of routine work, further 
emphasizes the importance of this principle. It explains at the same time Sir Ian 
Hamilton's by-law and certain cases where, in apparently successful organizations, 
larger numbers of subordinates report to a supervisor. It is obvious that, if it is cross 
and group relationships which introduce complexity into supervision, this factor will 
operate with much less force where the work done by each of various subordinates 


direct single a = 

cross b = 

direct group c = 

Fig. 7. Direct and Cross Relationships 

Table I Computed on Maximum Basis 

= n 

= n(n - 1) 




































total direct 
single and 

total direct 

total direct 
and cross 

d = a + b 

14 9 16 25 36 



81 100 



f = 

a + c 

a + b + c 

U 2 


1 4 12 32 80 192 448 1024 2304 5120 11264 24516 
1 6 18 44 100 222 490 1080 2376 5210 11374 24708 

direct single a 
cross b' 

direct group c' 

= 2 (n 


total direct 
single and 

total direct 
total direct 
and cross 

Number of 
per group 

d' = a + b' 

= 2" - n - 1 

2(n + l) 

Table II Computed on Minimum Basis 
12 3 4 5 
1 3 6 10 
1 4 11 26 

6 7 8 9 10 11 12 

15 21 28 36 45 55 66 

57 120 247 502 1013 2036 4083 


6 10 

15 21 







a + c' 

= 2" 

V = 

a + b' +c' = 2" +? ( n - 1) - 1 

13 7 15 31 63 127 255 511 1023 2047 4095 
1 4 10 21 41 78 148 283 547 1068 2102 4161 

Fig. 2. Chart Showing Direct Group Relationships 

1 Subordinate 

2 Subordinates 

1 A B 

2 AB 

3 Subordinates 

1 ABC 


3 ABC 

4 Subordinates 

1 A B C D 




5 Subordinates 

1 A B C D E 





6 Subordinates 

1 A B C D E F 






7 Subordinates 

1 A B C D E F G 











Note that for four subordinates it is quite easy to grasp and remember every combination of groups, but that from five on, this is no 
longer possible, because the various groups become a maze of confusion. 



does not come into contact with that done by others. This is frequently the case at 
the lowest level of organization, each worker being given an assigned task involving 
little or no contact with colleagues. On the other hand, at the higher levels of organi- 
zation where a large measure of responsibility and freedom is necessarily delegated 
to subordinates, themselves in charge of important divisions, the number and fre- 
quency of cross and group relationships is necessarily much increased. 

But even at these higher levels, where immediate subordinates are doing work 
which does not impinge upon that of their colleagues larger numbers can be con- 

Fig. 3. 

trolled. The General in command of a British infantry Division during the Great War 
had six subordinates reporting directly to him. Three of these were, however, in 
command of uniform infantry Brigades who had no regular cross relationships. Similar 
situations are found in business where a Head Office administers a number of sub- 
sidiary companies or branches widely separated geographically and having few re- 
lationships of a technical or functional character. But the general evolution of modern 
business tends towards an increasing degree of specialization. This enforces organiza- 
tion by function with a correspondingly greater demand for co-ordination, creating 
automatically a wider and wider range of group and cross relationships. The prin- 
ciple which has been discussed is, therefore, likely to become of greater rather than 
of less importance in all forms of organization. 





Administration has to do with getting things done; with the accomplishment of 
defined objectives. The science of administration is thus the system of knowledge 
whereby men may understand relationships, predict results, and influence outcomes in 
any situation where men are organized at work together for a common purpose. Public 
administration is that part of the science of administration which has to do with gov- 
ernment, and thus concerns itself primarily with the executive branch, where the work 
of government is done, though there are obviously administrative problems also in con- 
nection with the legislative and the judicial branches. Public administration is thus a 
division of political science, and one of the social sciences. 

At the present time administration is more an art than a science; in fact there are 
those who assert dogmatically that it can never be anything else. They draw no hope 
from the fact that metallurgy, for example, was completely an art several centuries 
before it became primarily a science and commenced its great forward strides after 
generations of intermittent advance and decline. 

It is fashionable for physicists, chemists and biologists who have achieved remark- 
able control and great predictive accuracy in narrow areas to ridicule all the social 
scientists, particularly those in government, because of the small body of verified 
knowledge thus far accumulated and "laws" formulated outside of the field of the 
"exact sciences." It is even denied that there can be any "science" in social affairs. 
This na'ive attitude is perhaps not to be wondered at in a group which has so recently 
arrived on the scene of man's intellectual theatre and has been permitted to play such a 
striking role for the past two centuries. Natural science, after all, has undertaken the 
comparatively simple and easy task of understanding the mechanistic and mathemat- 
ical relationships of the physical world and has left to philosophy, ethics, religion, edu- 
cation, sociology, political science, and other social sciences the truly difficult and the 
truly important aspects of life and knowledge. 

Social science rests at many points upon the physical sciences. Economics and 
politics, for example, are fundamentally conditioned by the discoveries of pure science 
and their technological application. The advance of certain of the exact sciences thus 
becomes of great concern and serves to condition and delay the advance of social sci- 
ence. The very fact that exact science has made a great forward surge in the past one 
hundred years, inevitably thrusts a whole new series of unsolved problems upon social 

The social sciences are also mutually dependent. This is particularly true of politi- 
cal science because political science is not a unitary, but a co-ordinating science which 
deals on one side with man's political life, desires and behavior, and on the other with 
government and public administration in which must be utilized most if not all of the 
professions and sciences which man has developed. 

The basic difference between the exact sciences and the social sciences, it has 
often been said, is that the social sciences must deal with values and ends. It is this 
which places them in a different category from the mechanistic sciences. It seems to the 



writer, however, that the importance of this element may be overemphasized, and that 
such overemphasis is certain to discourage the advance of social science as much as any 
other single factor. In many of the subsidiary but fundamental fields of social knowl- 
edge it is possible to put values and ends to one side, or to assume them as constants, 
just as is done in the pure sciences. For example, Gresham's law with regard to dear 
currency and cheap currency has validity entirely outside of any notion of what is 
"good" or "bad." Similarly Thorndike's studies in the age of learning, Boas' investiga- 
tion of skulls and culture, Huntington's weather records, Buck's examination of public 
budget systems, Hurd's study of the movement of land values, King's analysis of income 
distribution, Mitchell's investigations of the price cycle do not depend for their validity 
upon the "good" or the "bad." Value finds its place in these studies not in the state- 
ment of variations and interrelationships, but in the social appraisal and application of 
the principles deduced. This does not mean that the social scientist will not be led on in 
his quest for truth by his individual value-interests, but it does mean that the results of 
his work, if scientifically done, may be used by others who have entirely different 
values in view. It would hardly be claimed that the structure of American democracy 
profited more from the thinking of the democrat Jackson than from that of the aristo- 
crat Hamilton; nor that the concept of public and private law devised for the Roman 
Empire is of no value for a soviet republic. 

It thus behooves the student of administration, along with other students of social 
science, to acquire the habit of separating (a) relationships and (b) value judgments as 
far as is possible in his work. In scientific literature, at least, he should endeavor to say: 
"Under conditions x, y and z conduct A will produce B; and conduct A 1 will produce 
C." He may have discovered this because he feels that B is desirable and C is unde- 
sirable, but if another student, or a statesman, confronts the same problem he may 
none the less be able to build upon and make use of the scientific work of the first stu- 
dent even though he has a reversed scale of values. Whenever a student of government 
says: "The mayor should now do A," this is to be interpreted: 

1 . Present conditions are xyz 

2. Under conditions xyz, A gives B 

3. B is good, therefore 

4. Do A 

If political scientists will make it a habit to split up every important "should" sentence 
in this way, they will not only make more useful to others the ideas which they develop, 
but may also introduce into their own work a new element of scientific validity. In 
other words, "should" is a word political scientists should not use in scientific 
discussion ! 

In the science of administration, whether public or private, the basic "good" is 
efficiency. The fundamental objective of the science of administration is the accom- 
plishment of the work in hand with the least expenditure of man-power and materials. 
Efficiency is thus axiom number one in the value scale of administration. This brings 
administration into apparent conflict with certain elements of the value scale of poli- 
tics, whether we use that term in its scientific or in its popular sense.* But both public 

* See Frank W. Goodnow "Politics and Administration"; also "Politics and Administration" by the author in 
the Annals, September, 1933. 


administration and politics are branches of political science, so that we are in the end 
compelled to mitigate the pure concept of efficiency in the light of the value scale of 
politics and the social order. There are, for example, highly inefficient arrangements 
like citizen boards and small local governments which may be necessary in a democracy 
as educational devices. It has been argued also that the spoils system, which destroys 
efficiency in administration, is needed to maintain the political party, that the political 
party is needed to maintain the structure of government, and that without the struc- 
ture of government, administration itself will disappear. While this chain of causation 
has been disproved under certain conditions, it none the less illustrates the point that 
the principles of politics may seriously affect efficiency. Similarly in private business it 
is often true that the necessity for immediate profits growing from the system of private 
ownership may seriously interfere with the achievement of efficiency in practice. It 
does not seem to the writer, however, that these interferences with efficiency in any 
way eliminate efficiency as the fundamental value upon which the science of adminis- 
tration may be erected. They serve to condition and to complicate, but not to change 
the single ultimate test of value in administration. 

In other words, the student of administration must take into account the condi- 
tions under which a given group of men are brought together to do a job. These condi- 
tions may include not only physical obstacles but also the democratic dogma, the 
fascist structure, a socialist economy, or the spoils system. But in any case the student of 
administration will not only explore relationships from the standpoint of efficiency 
within the framework afforded, but will consider also the effect of that framework 
upon efficiency itself wherever the opportunity is presented. 

If it be true that the continual intrusion of varying scales of value has served to 
hinder the development of all of the social sciences, may it not be well to minimize this 
difficulty as is here suggested? This, it seems to the writer, is already possible in the study 
of public administration by regarding all value scales as environmental with the excep- 
tion of one — efficiency. In this way it may be possible to approximate more nearly 
the impersonal valueless world in which exact science has advanced with such success. 

But even so, great difficulties to scientific advance remain. If we may by various 
devices put fluctuating values to one side, and this is not as easy for other social 
sciences as it is for public administration, we are still confronted by two problems 
which the exact scientists have largely escaped. These are: 

First, in dealing with human beings we encounter a rare dynamic element which 
is compounded in unknown proportions of predictable and of unpredictable, of ra- 
tional and of emotional conduct, and 

Second, we are not able, except in the rarest circumstances, to set up controlled 
experiments or to test theories over and over at will. 

The human psyche is significant, not entirely because it is dynamic and in part 
unpredictable and irrational, but also because human beings are so extraordinarily 
rare. There are in one cubic centimeter of air 15,000,000,000 times as many molecules 
as there are individual humans on this earth. It is this scarcity of phenomena which 
makes the individual variations so difficult and important. If we had as many humans 
to deal with as the exact scientist has electrons, we might more easily discover the pat- 
tern of conduct and the normal probability curves of social life. And in political sci- 


ence, when we turn to aggregates of human beings, organized in nations, we are con- 
fronted by a situation of still greater scarcity. There is only one Soviet Union, one 
Great Britain, one United States of America. With this paucity of phenomena to ob- 
serve, it would be a miracle indeed if scholars were able to see through to the underly- 
ing laws and set them forth, certain that every significant variation was covered. This 
immensely important problem of variation, which is at the center of social science, was 
not even suspected to exist in the constitution of matter until a very few years ago, and 
even now presents a theoretical rather than a practical problem to the physicist because 
he, amid the plethora of phenomena, may rely on solid averages as a starting point. 

Social experiments, moreover, must be made by men on men. This greatly re- 
stricts the process of verification of hypotheses not only because of the value and dig- 
nity of human life, but also because human beings continually interfere with experi- 
ments involving themselves. 

There is no easy escape for social science from these two limitations. The number 
of human beings, though increasing, cannot remotely approach the gigantic statistical 
arrays which confront the physical scientist even within the confines of the smallest 
particle of matter. Nor may we follow the biologists and develop extensive controlled 
experiments to which human beings will readily submit over and over for the sake of 
pure science. Nor may we hope to develop laboratories in which outside social condi- 
tions may be reproduced for purposes of experimentation, for after all these labora- 
tories must contain active elements which behave just like human beings in a normal 
human setting — and this is precisely that which human beings cannot provide out- 
side of themselves. 

Should we look, then, to the invention of instruments as the open sesame of social 
science? Do we need for social science microscopes, or telescopes, or cathode-ray tubes 
— that is, instruments to extend our sensory equipment? Do we need thermometers, 
balances, barometers — that is, instruments with which to make more accurate meas- 
urements? Or should our search be directed primarily in some other direction? 

Though the writer has been greatly intrigued by the search for new instruments, 
useful particularly in public administration, and has contributed to the invention or 
development of some, * it does not seem to him that the invention of instruments for the 
extension and refinement of the senses is the prime necessity at the present juncture. It 
is not mechanical instruments we need to enable us to see that which now escapes us. 
The great need is putting ourselves in a position to use the instruments which we 
already have. 

What we require in the social sciences at the present time, it seems, is : 

1. Analysis of phenomena from which we may derive standard nomenclature, 
measurable elements, and rational concepts; 

2. The development of extensive scientific documentation based upon these 
analyses, and 

3. The encouragment of imaginative approach to social phenomena, and the 
publication and circulation of hypotheses so that they may be scrutinized by others in 
the light of experience, now and in future years. 

* The Merge-Calculator, the Proportional Representation Voting Machine, and various improvements in statisti- 
cal machinery. 


The analysis of phenomena, if it is to be of value in future years, or is to be added 
to the work of others and become part of a growing reservoir of knowledge, must be 
brought within a single system of definition and nomenclature. This is so obvious that 
it needs no further proof. It has been the device by which natural science has de- 
veloped, and makes it possible for each scholar to stand on the shoulders of his pred- 
ecessors, and not at ground level. 

Definition requires careful analysis, analysis which must include the dynamic as 
well as the static facts. This will in itself show the way to the elements which can be 
measured, translated to mathematical terms, and thus brought into such form that 
they may be subjected to the most complete system of logic and inference which man 
has created. In the development of meaningful measurements, there may be room for 
new instruments. But, here again, instruments are not the first need. The first need is to 
discover and name the things that are to be measured. Surely we already have in the 
punch card, the instantaneous electrical transmission of information, automatic ac- 
counting, the electrical scoring of examinations and schedules, the perfected "straw 
vote," the photo-electric cell, the cinema, the decimal system of filing and classifying, 
and similar well-known devices, the basic instrumental equipment which is necessary 
for the advance of the social sciences. We have barely begun to use these devices. It 
will be observed that they are useful primarily in the summarization of experience for 
analysis. This is precisely the process which is needed because in the social sciences we 
start with the restless electron, and endeavor to build up the solid continuum. 

The development of documentation is essential in the social sciences because it is 
the first step in accumulating sufficient data to submerge unimportant variables, and 
thus to furnish the basis of rational analysis. If we cannot have vast quantities of 
phenomena from which to work, we must at least accumulate those which we have 
from generation to generation so that scholars in considering the fate of mankind will 
not be confined to their own town and their own life span. The effort to "capture and 
record" administrative experience is surely fundamental. It is perhaps significant that 
modern science itself arose on the foundations of Greek analysis and documentation, 
and that science did not emerge even in civilizations further advanced than the Greek 
in some particulars, where such documentation was conspicuously absent. 

And how may we encourage the imaginative approach, the formulation of gener- 
alizations, the statement of hypotheses, the building up and testing of theories? There 
is, of course, no simple answer. But three things are certain : first, we must subsidize so- 
cial science research and philosophy through the universities and research institutes so 
that many men may be set free to study, think, and test out ideas; second, we must 
make it easy for those with ideas to secure their circulation among their fellows; and 
finally, we must contrive to give recognition to those who come forward with original 
and valid contributions. All of these factors played their part in the conquest of the 
natural world by exact science, and may be counted upon again to advance scientific 
knowledge and control in the world of human affairs. 


rt ^i i rop | igf^^vDate Due 

J3ate Due 


Withdrawn from UF. Surveyed te mm MSm 
Withdrawn from L -™ ■ eyed to internet Archive 

1 , i 

3 12b5 Dm3D DMM2 


''*»»i.,„„»««« ,,c